Methylation profiling of twenty four genes and the concordant methylation behaviours of nineteen genes that may contribute to hepatocellular carcinogenesis

Cell Research(2003)；1 3(5)：3 1 9--333 http：]Avwxv．cell—research．com

M ethylation profiling of twenty four genes and the concordant methylation

behaviours of nineteen genes that m ay contribute to hepat0cellular

carcln0genesls

JIAN YU ，HONG Yu ZHANG ，ZHEN ZHONG M A ，W EI LU 2 YI FEI W ANG 2 JINGDE ZHU

The State—Key Laboratory fo，．Oncogenes and Related Genes，Shanghai Cancer Institute，Shanghai Jiaotong University，LN 2200／25，Xietu Road,Shanghai 200032，China E—mail：

：Department ofMathematics，Shanghai University，No．99，Shangda Road,Shanghai 200436，China

ABSTRACT

To determine the possible role of the epigenetic mechanisms in carcinogenesis of the hepatocellular carcinoma，

we methylation—profiled the prom oter CpG islands of twenty four genes both in HCC tumors and the neighboring

non—cancerous tissues of twenty eight patients using the methylation—specific PCR(MSP)method in conjunction witl1 the DNA sequencing．In comparison with the norm alliver tissues from the healthy donors，it was fo und that

while remained unmethylated the ABL．CAV,EP0，GATA3，LKB l，NEP’NFL，NIS and p27m genes，varying

extents ofthe HCC specific hyperm ethylation were found associated with the AB0，AR，CSPG2，cyclin al，DBCCRl，

GALR2，IRF7，M GM T，M TlA，M Y0Dl，OCT6，p57mP2，p73，W Tl genes，and demethylation with the M AGEA l

gene．respectively．Judged by whether the hyperm ethylated occurred in HCC more frequently than in their neigh—

boring norm al tissues．the hyperm ethylation status of the AR，DBCCRl，IRF7，OCT6，and p73 genes was consid．

ered as the event specific to the late stage．while that the rest that lacked such a distinguished contrast．as the event

specific to the early stage of HCC carcinogenesis．Among all the clinical pathological parameters tested for the

association with．the hyperm ethylation of the cyclin a l gene was more prevalent in the non—cirrhosis group(P=O．

02 l while the hyperm ethylated P l 6 K4a gene was more common in the cirrhosis group(P=O．0 l 7)． The concor-

dant methylation behaviors ofnineteen genes．including the four previously studied and their association with cirrho—

sis has been evaluated by the best subgroup selection method．The data presented in this report would enable us to

shape our understanding of the mechanisms for the HCC specific loss of the epigenetic stability of the genome，as

well as the strategy of developing the novel robust methylation based diagnostic and prognostic tools．

Key words：promoter CpG island,methylation specific PCR，concordant behaviors of methylation

TRoDUCTIoN

Hepatocellular carcinoma(HCC)is one of the com— monest cancerous diseases rating the fifth in occur—

rence and the third in mortality worldwide『l 1．As it is

geographically biased toward the several parts of Asia

and Africa，China in particular．it presents one of the

major health threat in China[2，3]．The dismal prognos— tic future of the patients is largely attributive to the rap—

idly advancing nature and di伍 culties in early diagnosis

Of HCC．Therefo re．there are urgent needs for the ro一

‘Both authors contribute equally to this work．

Correspondence to：Jing De ZHU

Tel／Fax：00862 1—64224285；

E-mail：zhujingde@2 l cn．com or zhujingde@yahoo．corn

Received Aug-25-2003 Revised Sep一22-2003 Accepted Sep一23—2003

bust diagnostic，prognostic and even therapeutic ap—

proaches that can only be brought about by the much

improved understanding of the fu ndamental aspects at

various biological levels of the events during the malig-

nant transfo rm ation of the norm al hepatocytes． The

decades of intensive m olecular genetic analyses have

yielded a considerable amoun t of information on the po—

tential genetic defects associated with the natural cour se

ofcarcinogenesis ofhepatocytes[4—6]．Until recently，the epigenetic m echanisms without the changes in DNA

sequence has been found capable of profoundly affect—

ing the transcription status of both genes and repeti—

tively sequences，that subsequently confers the gr owing

advantage to tumor cells over their norm al counterparts

『7—91．The covalent addition of the methyl group at the

维普资讯 http://www.cqvip.com

Methylation profiling and concordant methylation behaviors in HCC http：／／www．cell—research．corn

C5 of cytosine in the CpG dinucleotide is essentially the

only form ofthe covalent m odification OfDNA in high

eukaryotes．having a number of biochemical as well as

biological implications It can eliminate the sequence

specific binding of the transcription factors to the cog—

nate eis—elem ents and promote the association of the

m ethyl CpG specific binding proteins to the methyl CpG，

with a cascade of reactions leading to the chrom atin

condensation and transcription silencing[10]．Over 50％ of the protein coding genes have at least one CpG island

within or near their promoters．Expressions ofthese genes

are subjected to the controls over the methylation state

ofthe CpG islan ds．Aberrant DNA methylation pattern

changes the gene transcription and has been etiologi—

cally linked to the occurrence of a number of genetic

diseases including cancers『lO】．The enzymes respon— sible for DNA methylation are the DNA methyl trans—

ferase I and IIIA as well as IIIB．The form er is m ainly

responsible fo r the m aintenance of the methylation sta—

tus of genomes after DNA replication，whereas the later

two act principally in the de novo DNA methylation in

the early development ofhigh eukaryotic organisms[1O]． E levated expression of these three methyl transferase

genes were reported in the majority of cancers tested， which may partly account for the increased local

hypermethylation『l O—l 21．However，recent evidences

demonstrated that the histone modification，methylation

of histone H3 in particular，might occur prior to the es—

tablishm ent of the DNA hypermethylation pattern that

contribute to the long．．term silencing of gene transcrip．．

tion[13]． Changes in the DNA m ethylation pattern s demon—

strated in all the can cers exam ined．consist of the global

level hypom ethylation in parallel with the local

hypermethylation[1 2，14．The genome—wide hypome— thylation can result in active transcription of the

tran sposon like repetitive sequences(such as the Alu

andLINE repeatsinmammals)thathavebeenlinkedto the increased genome instability，a predominant hallmark

of can cer cells『l 5一l 71．The hypermethylated status of

the prom oter CpG islands has been link ed to the ex—

presslon silencing or the tumor suppressor genes and

implicated as the 2nO hit，reminiscent to the loss of het—

erozygosity or other type of genetic deletion for total

inactivation of the tu mor suppressor genes in cancers

[1 8—2 1 1．TIle loss of the genetic imprinting attributed to the chan ges Of DNA methylation．such as reactivation

ofthe IGF一2 gene has been linked to the rapid prolifera一

320

tion oftumor cells in several types ofhuman tumors[22]． The reverse process，i．e．，demethylatin can also result in

the transcription activation of the otherwise inert genes，

including c-myc，and c-ras ，even though all ofthese genes

lack the typical CpG island witllin or near to the promot-

ers[23]．The association between the hypomethylation of the promoter CpG island and over-expression state of

the genes such as SURVIVIN[24】and hTERC that en—

codes the RNA component of the telomerase[25]have

been recently reported．

M ethylation profiling of the prom oter CpG islands of

theknowngeneshas beenan importantinform ationgath—

ering process for new insights in our understanding of

the mechanism s of the DNA methylation in both initia—

tion and progression of the carcinogenesis，as well as

the new clues for development of the relevant diagnos—

tic and prognostic m ethods and even for therapeutics

against cancers．The recent collective efforts have iden—

tiffed a list Of over one hundred genes the prom oter

CpG islands of which change in various tumors(http：／／

、)lr、v、)lr．missouri．edu／-hypermet／list ．．_

of promoters htm)

However，as the majority ofthe studies to date had only

targeted one or a few genes in rather small patient groups，

the concurrent hypermethylation behavior of multiple

genes has only been addressed in few tumor types，ex—

ceptforveryfewexamples[26]．Priortoour recentwork

『271 where methylation profile of the promoter CpG is—

landsoftw entygenesin tw entynineHCCpatientsamples

were presented，there had been only two reports de—

scribing the HCC specific chan ge in the m ethylation pro—

filing of three tumor suppressor genes：p l 6 柏．cyclin—

dependent kinase inhibitor 4b(p l 5INK4b)and the alterna— tive reading flame of the cyclin—dependent kinase inhibi—

tor 4a(D 1 4Ar~ )，respectively．In that study we found

that sixteen genes adenomatosis polyposis coli(APC)，

apoptotic protease activating factor(APAF 1)，breast

cancer 1(BRCA 1)，cadherin type 1(CDH 1)，death—as—

sociated protein kinase l HCC(DAPK l 1，mutL ho—

molog 1(hMLH l 1，Telomerase RNA component

(hTERC)，p 14ARF，p 1 5。伯，phosphatase and tensin ho—

molog(PTEN1，ras association domain fam ily l protein

isoform l c(RASSF 1 c1，retinoblastoma l fRB 1)，retinoic

acid receptor,beta(RAR—D)，SURVⅣIN，tissue inhibi—

tor of metaUoDroteinase 3(TIMP3)and von Hippel—

Lindau syndrome fVHL、remained unmethylated in all

the sample tested，whereas the following four genes：

Caspase 8(CASP8)，H—cadherin(CDH l 3)，P l 6INK4

and RASSF l a genes displayed the HCC associated


Cell Research 1 3(5)，Oct，2003 JianYU，HongYuZHANG，etal

hypermethylation to varying extents．The lack of the

HCC specific changes in the sixteen of the twenty genes

whose hypermethylation state of the promoter CpG is—

land has been linked to many other types of cancer[27] was indeed a surprise．In order to establish the concor—

dant methylation behavior of the genes displaying the

HCC specific changes．we further extended our study

to other tw enty four genes to assess the extent of the

methylation mediated mechanisms in the HCC an d found

fifleen genes had displayed the HCC specific changes．

By the stringent mathem atic analyses of the concordant

methylation behaviors of the nineteen genes(including

the four in the previous study[27])．the subsets of the

tw o to nine genes have been established in HCC and its

cirrhosis／non—cirrhosis subgroups，which may provide the

useful clues for the DNA methylation based diagnostic

and prognostic assays for HCC．

IATERIAI S AND M ETHoDS

Tissue samples andDNA extraction

w ith the informed consent of all patients and donors and ap．

proval of the ethics committee．the samples of tumor and adiacent

non．cancerous tissues were collected from HCC patients rn=28)

during surgery at The Qidong County Hospital，The Oriental Insti—

tute for Liver Diseases an d Guan gxi Provincial Hospita1．respectively．

In addition，normalliver tissues(n=4)were obtained from liver

donors at the Liver TransDlantatiOn Unit in The First Affi liated

Hospital，College of Medicine，Zhejiang University．The pathologi— cal classification OfHCC tissues was carried out and the stage ofeach

HCC was determined according to criteria outlined by the Liver

Cancer Study Group of Japan[281．Special efforts were made to

include the corresponding non—cancerous tissues．samples of which

were overlooked in all previous studies of HCC[24]．Furtherm ore． pathological examination ofthe tissues had been carried out to elimi．

hate samples in which contamination ofundesirable tissues exceeded

20％．Tb emphasize the HCC specificity of this study,norm alliver

tissues from four male healthy donors were collected from the Liver

TransDlantation Unit of the Zhejiang First Hospital as the norm al liver contro1．

Total genomic DNA was extracted from frozen tissue specimens

(50一lOOrng)according to standard protocol with some modifications

『271．Frozen pulverized powders of the specimens were re．SUS． pended with 2 mllysis buffer：50 mM Tris．HCl pH 8．0，50 rnM

EDTA．1％ SDS．10mM NaClplus 100mg／mlboiling．treatedRNase

A(Sigma)．Following one hour of incubation at 37 C．Proteinase K

(Roche．USA)was added to the celhlar lysates for a final concentra． tionofl00me,／mlan dthedigestionwas carriedout at 55 Cfor2h．

Organic extractions with a half volume of Phenol／Chloroform／

Isoamyl alcohol f l：l：0．04)were repeatedly carried out until no vis．．ble interphase remained after centrifugation．DNA was precipitated

from the aqueousphaseinthepresenceof 0I3M NaOAc oH 7．0an d

two and a half volumes of ethano1．The DNA pellet was washed

oncewith 70％ ethanolan ddissolved at65℃ for 30minwith 0．2．

0．4 ml TE(10 mM Tris—HC1 pH 7．4 and 1 rrvl4EDTA)，followed by storage at 4。C until further use

． The DNA concentrations were

calculated according to their OD260 readings．

Bisulphate treatment ofDNA and methylation spe—

cific PCR(MSP) The primer pairs for the methylation specific PCR in this report

were either adopted or designed accordhag to the same principle wim

assistance of the software packages for the CpG islands identifica—

tion(http：／／www．uscnorris．com／cpgislands)and the primer design

(http：／／micro—gen．ouhsc．edu／cgi—bin／primer3_

www．cgi)．

The methylation status of the promoter CpG islands of tw enty

four genes in all sample DNA were an alyzed by M SP on the sodium．

bisulfite converted DNA『27]．Indetail，l0mgDNAin 50mlTEwas

incubated with 5．5mlof3M NaOH at37 oCfor l0min．followedby

a l 6 hour treatment at 50 oC after adding 30 ml of freshly prepared

l 0 mM hydroquinone an d 520 ml of freshly prepared 3．6 M sodium ．

bisulfite at PH 5．0．The DNA was desalted using a home—made

dialysis system with 1％ agarose(detailed protocol will be provided

upon request)．The DNA in the desalted sample fapproximately 100

mlinvolum e)wasdenatured at 37。Cfor l5minwith 5．5mlof3M

NaOH followed by ethanol precipitation with 33 ml l0 ̂NH．0AC

and 300 ml ethano1．Atier washing with 70％ ethano1．the gently

dried DNA pellet was dissolved with 30 ml TE at 65 oC for l0 min．

The DNA sample was finally stored at-20 C until further use．

PCR reaction was carried out in a volum e of l5 ml with 50 n2 or less

template DNA with FastStart Taq polymerase(Roche，Germ any)as

follows．Atier an initial heat denaturing step 4 min treatment at

94oC．30 cycles Of 92。C for l 5 sec．varying temperatures with primer

pairs for l5 sec and 72 oC for 20 sec．was carried out．The PCR

products were separated by 1．2％ ethidium bromide containing aga．

rose gel electrODhOresis with l X TAE and visualized under UV

illumination．To verify the PCR results．representative bands from

each target were ge1．purified and cloned into T．vector(Promega．

USA1 followed by automatic DNA sequencing provided by BoCai

(Shanghai．China)．Only verified results were presented in this report．

Statistics

The methylation data were dichotomized as 1 for the co．existence

of the methylated and unlTlethylated alleles，2，for methvlated allele

onlv and 0 f．or me unmethvlated f．or bom aUeles to f．acilitate statisti．

cal analvsis usin2 contingencv tables．The memvlation profiles of

each individual gene(in percentage)classified by me cirrhosis status Ofthe patients were presented bOm in table and in plOt．The statistic

analVses for the assOciatiOn between the memVlatiOn profile Of me

gene and each Oft}Ie clinica1．patllOlOgical parameters were c ed out

with me statistics package(htcp：／／www．R—project．or )，where bom

Pearsong’s Chi．square test with Upton’s adiusnnent and Fisher ex．

act test(http：／／www．R—project．org／)used to deal with the sample cells with the low expected values． The relative frequencv wim a

95％ confidence inter、，al(P<0．051 for a binomial distribution was

calculated f_0r me whOle as well as each sub ，pes Of astrOcvtOma

patients．

The cOncOrdant memVlatiOn behaviOrs Of me 2enes were estab．

1ished by comDaring me relevant Occun．ence OfvariOus subsets cOn．

taining two，thr nd four genes respectiVely，with me best sub—set

selection method『291．

321


Methylation profiling and concordant methylation behaviors in HCC http：／／www．cell—research．com

G ene Nanle GeneBank N o

ABL(M)

ABL(U)

ABo(M)

ABO (U)

AR(M)

AR(U)

CAV(M)

CAV (U)

CSPG2(M)

CSPG2(U)

cyclin al(M)

cyclin al(U)

DBCCR1(M)

DBCCR1(U)

EPo (M)

EPo (U)

GALR2(M)

GALR2(U)

GATA3(M)

GA1_A3(U)

IRF 7(M)

IRF 7(U)

LKB1(M)

LKB1(U)

U07563

U22302

M 581 58

N T 007933

N T 007022

AF1 24143

A L1 38894

NT 007933

NT 010641

N T 033983

N T 008953 8

XM 01 291 3

M AGEA1 fM 、U82670

M AG EA 1 fU、

MGMT(M、 AL35553 1

MGMT(U1

MT1A fM、 K01 383

M T1A fU 、

M YoD 1 fM 、 AF027148

M YoD1 fU、

N EP fM 、 X 79438

NEP fU、

NFL fM、 X05608

N FLfU 、

N IS fM 、 A C005796

NIS fU、

oCT6 fM 、 L26494

oCT6fU、

p27 (M) AB003688

p27 (U1

p57 (M) HSU48869

p57 (U)

p73(M) AB031 234

p73(U)

W T1fM 、 X 74840

WT1 fU、

Tab 1．The target promoter CpG islands and the primers for MSP

Sense 5’．3

CGAGGATCG TTATTGGTTCG

TTGTTATTGGTTTGGGTTGG

TTAAGG TATTAGGGTTACGA GGGGC

GG ATAGGG TTTTAAGG TATTAG GGTT

TTTCG AGATTTCGGG GAG

TTTTGAGATTTTGGGGAGTTAG

TTAGGG CGAG AGCG ATTC

AGTGTTTAGGGTGAGAGTGATTTG

AGTTTCGG GGG ACG TTT

GAGTTTTGG GGGATGTTTTT

TCGTCGCGTTTTAGTCGT

GGGTAGTTTTGTTGTGTTTTAGTTG

CGGGTGTAGCGTTTCG TA

GGTGGTGTTGG GTG TAGTG

TTTACGTTTCGGCGAGTT

11TTGTT1_ATGTT下TGGTG AGTTTTA

GTTTCGGCGCGTATTTTA

TGGAGTGGTTTTGGTGTGTAT

GGTCGTACG TCGTCGTTT

GATTGA GTAGGTTG TATGTTGTTG TTT

G TTTCGCGG AGTTG AG AATC

GGTGGGGTTTTG TGG AGT

GGTGTTCGTCG GTTCG TA

TTTGG GTTTGTTGGTTTGTA

GTTCGGTCG AA GGA ATTTG A

GTTTG GTTGAA GGAATTTGA

AGCGTCGTTGTTTTG TGC

TTGGTAGTGTTGTTGTTTTGTGT

TAAG GTTGGGTTTTCGG AAC

TAAG GTTG GGTTTTTGG AAT

GACGG TTTTCGA CGG TTT

ATTTGATGGTTTTTGATGG TTT

GGTGTTTCG TCGTTTTACG

GTTTAGGTGTTTTGTTGTTTTATGG

TTTATTGGTCGG CGTGTC

TGTTTTTATTGGTTGGTGTGTT

G TTTCG TTGCGGTCGT

ATGTTTTGTTGTGGTTGTGGT

ACGAG GATGCGTTTAG TTCG

GGATGAG GATGTG TTTAGTTTG

CGACGTCGGTAAGGTTTG

TGTGATTTTGATGTTGGTAAGGT

GGG TTCGCGCGTATAA A

GGG TGG GGTTTGTGTG TAT

GCGTTCGGTTCGTAGGTT

TG GGTGTTTGG TTTGTAGGT

G TTAG GCG TCGTCG AGG TTA

TGGGATTTGGG TGGTATTTG

RESUI S AND DISCUSSIONS

Clinical-pathological considerations

The g

cally in

eographical distribution of HCC varies dramati—

China．In areas such as Qidong county near

Shanghai as well as Guangxi province，the annual HCC

incidenceisashi gh as70—96／10，0000andapproximately

sevento eightfoldhigherthanthatin thelowincidence

areas in China[30]．In view ofthe high likelihood ofthe

322

Antisense 5 3’ Genomie position Size(bp)

TA ACCG AAACGCGCCTATTA

TA ACCACAACCACATCTCCAC

CGACCATAACTCCGCGTCT

CCACATCTAATCTCAACCTCCA

ACAA CTCCG AACG ACGA C

CCAAA CA ACA ACTTCAA AACCA

GCCG CCA AA AATCA AAA C

CCACCCACCACCAAAAAT

TTTTC1lACCCCGCTCTCC

AACACCCAAACCACTCCA

ACCCGTTCTCCCAACAAC

AACCACTAACAACCCCCTCT

CA AAAACCCCCTCCCTAA

CA AAAACCCCCTCCCTAA

CG AA CG ACCG AAATAA CC

A CCCCAATCCAA CTCCAA

CACGCAA AATCCGA AATC

CAAAATCAA CCCCTCCA C

AAA ACA CCG ATCCCG AAA

CCA ATCCCA AAACCATCC

TATAACCGACGCGC ACAC

TACA AATATAACCAA CACACACAC

TTCCGACTTCCCTTCTCC

TA．rlATTCCAACTTCCCTTCTCC

CCACAACCCTCCCTCTTAAA

ACCCACAACCCTCCCTCTTA

CGCTTTCA AAACCACTCG

CATCCTACAACCCCCACA

AAATACGAACCACGAAACCA

CTCCCCTAA ATACAA ACCACA

G CCCGA AACCGA ATACAC

CACA CACATACTCATCCTCACA

CATC CCG ACCAATAAACG

CCAACACATCCCAACCA AT

TACTCCCCGACGACG AT

ACATCTCCACATAACACCACTT

TACCGCACG TCCATTAACT

ACACATCCATTAA CTTCTCTACCC

GAATCGATCTCCTCCAACCA

TCAATCTCCTCCA ACCACTT

A AACG CGCA AAA ACTAC G

CA AACCACA ACCCA AACTCT

ATACGAAAAACGCGACGA

A AACACAACA ACTACCTAACTATC

CTCAACTCCCAAAACCCAA

CCAA CTCTCAA CTCCCAAAA

AAAACGCAAAATCCAACACC

CACCAACACCCACTACA CCA

- 368 to -172 197

- 362 to -150 21 3

-243 to+6 250

- 253 to -5 249

-3 tO +210 214

．3 to +204 208

+1320 tO +1482 163

+1315 tO +1488 1 74

+1351 tO +1 526 1 77

+1350 tO +1501 1 52

． 755 to -550 206

- 762 to -565 1 99

+396 to +561 1 66

+388 to +561 1 74

． 212 tO-28 1 85

．217 tO +1 7 236

-400 to -246 1 55

． 389 to -239 1 51

-659 to -468 1 92

-668 to -474 1 95

+1617to +356 196

+155 tO +362 208

- 253 to -68 1 68

- 238 to -64 1 75

+247to+345 322

+24 to +347 324

．451 to -266 1 86

．469 to -261 209

-421 tO-258 164

421 tO -251 1 71

+210 to +393 1 84

+206 to +41 8 21 3

+35927to +3760 1 69

+3587 tO +3766 1 80

-81 tO +81 163

-85 tO +159 245

- 12279 to -12115 1 65

- 12281 tO -12119 1 63

+727 to +955 229

+725 to +952 22 8

- 355 tO -163 193

- 363 to -l4l 223

- 70 to +118 205

- 79 to +111 203

- 1722 tO -l 511 21 2

- 1725 tO -1 505 221

+32l tO +526 206

+295 to +510 21 6

potential geographic impact，we deliberately recruited

nineteen patients in the Shanghai area，including eight

patients from Qidong County，and nine patients from

Guangxiprovince．A11thepatientswerehepatitisB vi—

rus fHBV)infected by both imlTlUnO—serological assays and the PCR test for existence of the HBVX gene in

tumor tissue(result not shown)．No geographical varia— tion is detected between these tw o patient groups．For

instance，male patients accounted for 78．9％ and 80％，

cirrhosis occurred in 52．6％ an d 50％ ofpatients and


Cell Research 1 3(5)，Oct，2003 JianYU，HongYuZHANG，etal

diagnoses of grade 1 were 28％ and 30％，gr ade I1 were

44％ and 40％ and gr ade II1 were 28％ and 30％ ofthe

Shanghai and Guangxi patient groups，respectiVely[27]． Therefore，the remaining analyses were carried out with—

out any consideration of geogr aphic impact．

Methylation profiling

The methylation specific PCR(MSP)on the bisulphate

treated DNA has been widely used for its simplicity as

well as speed．In the previous work．we verified the

M SP data by sequencing the representative PCR prod—

ucts and validated the data of two among tw enty genes

studied with a non PCR m ediated method，f．e．，South—

em analyses ofthe DNA digests by the methylation sen—

sitive enzymes[27]．In this extended study，therefore，

the MSP in conjunction with the DNA sequencing ofthe

batch—treatedgenomicDNA from the samepatientgroup

and the norm al control was adopted．

Th e in vivo malignant transfo rm ation of norm al cells

is a multiple—stage process，involving at least a dozen of

genetic and epigenetic changes．In this connection，there

is a prevalent notion that the morphologically“norm al’’

cells adjacent to the cancerous tissues may have already suffered from the genetic or／and epigenetic changes that

are specific to the early phase of m alignant

transform ation． To take this favorable assumption into

accoun t，we deliberately recruited，both in this report

and the previous．the neighboring non—cancerous liver

tissues in addition to HCC tissues into study，which have

been often excluded in other similar studies(J271 and

references within)．

Th e concerns were fully iustified as to the cross—con—

tamination of the norm al tissue in the HCC samples．or

vice versa． However，this seemed unfounded in our

studies．First，over forty four genes(twenty four genes

were in this study and tw enty genes were in the previ—

OUS report[27])．were chemically treated in batch and methylated profiled in group． Th e obtained well diversi—

fled methylation patterns of these forty four genes are

lly incompatible with the assum ption that there may be

the severe cross．．contamination of the undesirable tis．．

sues in the design ated samples．Second．M SP is a sen—

sitive m ethod capable of detecting a very low level of

contamination ofthe un desirable DNA．Th irdly．in addi—

tion to the sequencing verification，the HCC specific

methylation profile ofthe p14A~ an d p15rNK4 genes had

been confirm ed by a non—PCR based Southern analysis

of the digested DNA with the m ethylation sensitive re—

striction enzymes[271．

Another issue of im portance is concern ing the me—

thylation heterozygosity in any given samples．As shown

the Fig l and Tab 2．although some samples were ho—

mozygously methylated or unmethylated，the majority had the co—existed methylated and unm ethylated alleles．In—

clusion of the norm al liver tissues in our study enabled

us to establish the norm al methylation pattem of all the

forty four genes，that except for the MAGEAl[3 l】be— ing hom ozygously methylated，the rest forty three were

homozygously demethylated[27】(Fig l and Tab 2)．

Therefore，any changes in the methylation pattern from

the norm al liver tissue’s to the non—cancerous neighbor—

ing tissues or HCC tissues should indicate the involve—

ment of the DNA methylation m ediated events during

the HCC carcinOgenesis，that have been scored as the

positive methylation change．讳 ether the alleles lack．

ingin HCC specificmethylationchan gemayloseitsfunc—

tion via the genetic mechanisms remains to be addressed

in the futu re．

The in vivo m align ant transform ation of norm al cells

is a multiple—stage process，which involves multiple ge—

netic and epigenetic changes．In this conn ection．it has

been generally accepted that the morphologically‘'nor-

mal”cells adjacent to the cancerous tissues may have already been abnorm al at both genetic and epigenetic

levels，reflecting the early—phase—specific alteration of

the malign ant transform ation．W e．hence．recruited the

neighboring non—cancerous liver in addition to HCC

tissues，to address the stage—specific natu re of the me—

thylation changes that are likely to be reflected in the

neighboring non—cancerous and HCC tissues，respec．

tively．

Methylationprofiling oflwentyfourgenes with or

without the proven roles in carcinogenesis ofthe

human cancers

The consequence of the tumor-specific defects in the

epigenetic homeostasis is global，and some detectable

changes may have etiological role to play whereas oth—

ers m ay simply the by—stander by nature．Therefore，

from thelistofthepromoterCpG islan dcontaininggenes

(http：／／www．missouri．edu／-hyperm et／list of promoters．

htm)we deliberately selected the targets that lack any

proven etiological role in carcinogenesis of other tissue

323


Methylation profiling and concordant methylation behaviors in HCC http ：llwww．cell-research．com

origins，in addition to those do．The form er group con—

sisted of the genes encoding the growth factor for the

erythropoiesis(EPO)【32]，a ubiquitously expressed tran-

scription factor(OCT6)[33]，the blood cell typing anti-

gen(ABO)【34]，and the myogenetic or erythropoietic

lineage—specific transcription factors(MYOD 1 of

GATA3)『351『361as well as the light chain of

neurofilarnent(NFL)[37]．Inthe secondgroup，therewere four tunlor suppressor genes，including two cyclin-de—

pendent kinase inhibitors：p27 【38]and p57KIP2【39]；

p53 analogue，p73138，40]，as well as the Wilms tUlrlOr 1 gene．W T 1． There were seven genes encoding the

surface proteins or nuclear receptors acting actively in

the intercellular interactions：galanin receptor 2(GALR2)

【42]，melanoma specific antigen A 1(MAGEA 1)，the

membrane metallo—endopeptidase(NEP)『43]，solute

careerfamily 5(NIS)【44]，cax,eolin 1(CAV)【45]，chon-

droitin sulfate proteoglycan 2(CSPG2)【46]and andro—

gen receptor(AR)【47]．Three genes implicated in sig-

nal transduction，cyclin a l『48]，the interferon regula．

tory factor 7(IRF7)．and a serine／threonine kinase 11

(Peutz—Jeghers syndrome)gene(LKB 1)【l 8]were

selected．The proto—oncogenes in this group were：the

gene encoding v—abl homologue 1(ABL)[49]and for the deleted in bladder can cer chromosome region candi．

date l(DBCCR1)『501．The final two in the list were

the genes that may be responsible for detoxification of

liver cells：O一6一m ethylguanine—DNA methyltransferase

0VIGMT)【18]andmetallothionein 1 A gene(MT1A)[51]．

As shown in Fig l and Tlab 2． within the group of

genes maintaining unmethylated in all three types of

samples tested，there are the genes devoid of any dem ．

onstrated association with tumors：EPO(panel 8 ．GATA3

(panel 10)，NFL(panel 18)andNIS(panel 19)，as well as the geneshavingthedemonstratedhyperm ethylationme．

diated gene silencing in the human tumors of the non．

HCC origins：ABL(panel 1)，CAV(I：Ianel 4)，LKB1(panel

12)，NEP(panel 17)，and p27KIPI(panel 21)genes．The simplistic explan ation would be that either inactivation

ofthese genes in HCC via the genetic based mechanisms，

or HCC form ation does not require the functional inacti．

vation of these genes．

Thegenes with theHCCspecifically alteredmethy—

lationprofiles

Fourteen genes：ABO(panel 2)，AR(panel 3)，CSPG2

(panel 5)，cyclin al(panel 6)，DBCCR1(panel 7)，GALR2

324

(panel 9)，IRF7(panel 11)，MGMT(panel 1 4)，MT 1 A

(panel 15)，MYOD1(panel 16)，OCT6(panel 20)，p57

(panel 22)，p73(panel 23)，and WT l(panel 24)were

unmethylated in all four cases ofthe norm al liver tissues．

an d methylated to various extents in the patient’s samples．

Among them ，there were the genes devoid of any obvi—

OUS tumor association：1．ABO(transferase A for the

ABO blood typing)，2，MYOD l(the myogenic specific

transcription factor)，and 3，OCT6(a common transcrip—

tion factor with POU domain)．The hyperm ethylation

frequency was as higher as 32％(9／28)and 50％f 1 4／28)

for ABO gene，54％(15／28)and 54％(15／28)for

MYOD 1．as well as 50％(14／28)and 82％(23／28)for OCT6 gene in the non—cancerous neighboring liver and

HCC tissue．respectively．Despite the higher incidence

of changes in the methylation profile．it is not possible to

discard the possible by．stander nature of such alterations

(Fig 1 and Tab 2、．There were suggestions for the possible aging related increase in the prom oter CpG island

of the genes including OCT6 and MYOD 1『52，53]，

However，it seem s not the case in this stu dy as there

was no obvious correlation betw een the age ofth e HCC

patients and the occurrence of the methylated status of

the prom oter CpG island．Alternatively,likely loss of

these three gene expression(extrapolated from the

hyperm ethylated status of the relevant promoter CpG

islands)indeedplay somepartsintheHCCform ationin

the unknown manners，which may deserve the further

investigation．

The rest are the genes im plicating in the human tu ．

mors ofother tissue origins via the methylation mediated

gene silencing(http：／／www．missouri．edu／～hyperm et／

list of promoters．htm)：AR(panel 3)，CSPG2(panel 5)，

cyclin al(panel 6)，DBCCR1(panel 7)，GALR2(panel 9)，

IRF7(panel l1)，MGMT(panel 14)，MT1A(panel 15)，

p57KIP2(panel 22)，p73(panel 23)，and WT l(panel 24))

or demethylation mediated activation：MAGEA 1(panel

1 3)．Among these targets，the MGM T gene was rarely

methylated(3．57％，l／28)．It encodes the enzyme ca．

pable of removing the alkylating adducts from the Of6

position ofguanine an d protects the cells from cytotoxic

and mutagenic effects f201．The MT lA gene that en．

codes the protein responsible for the cell detoxification

targetingatvariousadversary stimuliincludingtheheavy

m etals．has been reported being inactivated by the

hyperm ethylated promoter CpG island in rat hepatoma

【5 1]．It was only marginally hyperm ethylated in HCC

(10．71％，3／28)，indicatingthattheDNAmethylationme—


JianYU，HongYuZHANG，etal

Pand 1 Panel2

ABL ABO

o 哪 o o 删 a幅 C o ●_ o ●●C Q ●州 o●●C D lll~ 州 ∞ D圳。奢．c D — 一 o●∞ Z

Panel3

AR

∞ ●一。奢．c D— 一 o ●∞ Z删 Z 猫 H Z0摹C

蓑：：：嚣：：：：?：：：：?：麓：：I- ： ‘：：：

Panel7

D IN o 'C O14#4 11~ 4G D15̈ ‘H ●C o'州 o'∞

嚣：：： T T：。T：：：：：：嚣：：：：：：A T ⋯QQ Q⋯Q Q T ⋯~： ==== T G G ·-_-_：：：：：二：二：：二：：!

⋯ ：： !：!!工：：：：tt!!：c c!⋯ ⋯ t

Panel9

C JR2

H_ c_ ．删_ ．哪_

Panel 11

m F7

H_ c_ ．

：：》= 暑：￡嚣⋯"roe⋯QA T：霉事飘

!I!! 连 ! ±t!! 墨 ! §o !! t!! ! § Tc 再

Panel 13

^tAG EA l

吣 _ ． H_

Panel4

c

o圳 o o●_ 嘲 C a ●■■ o ●●C a ●■■ o●●C

U _ U _ U _ U _ · U _ U _ U _ u _

啊‘嘲_ ‘。．矗 Ⅲ -．_ ‘ 品⋯ ⋯ ‘气暑 ⋯ 川 ‘ ¨ ⋯ ‘ lT-‘一

：= ．矗 6o 筮：；：；赫；：靥；i ̂ ：；：：：

Panel6

cycnn al

O41N O41C a圳 a|．c a删 ● ．，c o 。嘲

Panel8

o 圳 o a●删 a a ●_ o●●C a ●_ o ●．c

U _ U _ U _ U _ U _ U _ U _ U _

_ ■ ●q■0⋯ ‘ ㈨ ⋯ 制 ¨ k ● ^●●● ⋯ ⋯ ^●●● ●r ⋯ ●⋯ ¨ Ⅲ T ． ^⋯ -． -

}⋯ ” t ⋯ ‘ e TTT uTT ¨G．TⅢ tt eⅢ TTTt cc ●-

■■●_ ^̂ ^^，h ^̂ 、̂ ．̂̂ ^̂ ^．一^^̂ ^̂ ^^ ，̂，Uu — JL ⋯ J ^^^̂ —^̂ ^·A—U0— ̂ ^

Panel 10

Panel 12

o圳 o o●_ 口— C a●■■ o●●C a ●■■ o ●●C

Panel 14

M G ^仃

．喇_ 。c_ ．_ 7c_

325


Methylation profiling and concordant methylation behaviors in HCC http：／／www．cell—research．com

Panel 15

仃̂ 1A,

a一 ● '~4tN '~4tC o删 o ．7c o ●删 o●婶

● ● ● ⋯ ● ⋯ ．．． ⋯ ● ⋯ ⋯ 川 ⋯ ● ⋯ ⋯ ⋯ ● ^ ̂

Pariel 17

●n

a ●●■ a●●C o●●一 o ●●C a●删 a ●●C ~ I4BN a ●●C

P ariel 19

N 珞

a● ●一 a ●●C o ●●一 o●●C a‘删 a ●●c a‘删 a ‘●c

兰= ：蒂 = ，一 - p、̂『T1 nrrrvWⅥ rrn、n ，̂nfv、 nf、，1fr wY、r

P ariel21

p27

o●翻 o●2C o●州 o ●3C a■●一 a ●●C o●5N Q#tff,C

P ariel23

p73

0’●一 0’·●C l''●H D1●C D1●H D1●C Z■2N Z∞ C

P anel 16

M Y 0 D 1

D1州 D1eC D2州 D2●C D■州 D ●●C Z■ Z∞ C

⋯ ” 泉 ⋯ ⋯ T Ⅲ ⋯ T毕rⅢ ⋯ ̂ ⋯ ⋯ T{器⋯ ⋯ ▲ ⋯ ⋯ c{替

P anel 18

N FL

o ●●一 a●|c o ●●一 o●●c a—●N a●●c a●●N a●．c

0 CTl6

P anel20

D1N D1C ■删 Z州 Z柚 C a，'一 o

～昌I 1=； a：；。；黼▲n：；；̂ 矗 ▲̂ ；

P ariel22

p57

． ”H“ ．。·cu ”H．．口。-三．．u =．2Hu

_ ．． ⋯ - ． Ⅲ ^ T - ⋯ ● ● ●⋯ ^ ● ●● ● ● 啊 T ．．． ⋯ ⋯ ⋯ ⋯ ⋯

P anel24

w T 1

D1●N D1eC D1删 D1●C D17H D1"Fc a 删 a 4TC

Fig 1． Methylation profiles of the promoter CpG islands of twenty four genes in HCC．Both electrophoretic patterns of the

representative PCR products ofeach oftwenty four targets(indicated respectively，at the top offigures)and the sequencing verification ofthe one representative PCR product were presented．To indicate the methylation status．the sequenced data are aligned with the wild-

type sequence．，sizemarkers，thebandsof250bpan d100bpwere shown．U，theunmethylated ；M ，thehypermethylated ．pan els：1，

ABL；2，ABo；3，AR；4，CAV；5，CPSG2；6，cyclin al；7，DBCCR1；8，EPO；9，GALR2；10，G A3；l1，IRF7；12，LKB1；13，

M AGEA1；14，MGMT；15，M T1A；16，MYOD1；17，NEP；18，NFL；19，NIS；20，OCT6；24，21，p27KIP’；22，p57Ⅺ ；23，p73 and 24，

WT 】．

diatedmechanism shouldnotplayany significantrolein

its inactivation，ifthere is any，in HCC．The following

three genes：CSPG2(1 7．86％)，GALR2(2 1．43％)and

p57K啦rl4．29％)were moderately hypermethylated in

HCC．expression of which may not be significantly af-

fected by the DNA hypermethylation of their promoter

()DG islandinHCC．

TheAR gene encodestheandrogen rec印 torthatplay

326

akeyrole inth e sign al transductionpathways ofcells to

respond to the male steroid hormone，androgen and was

reported to be inactivated via the epigenetic mechanism

in prostate cancers[54]．HCC seems an androgen—de—

pendenttulnorasitoccursfivetimesmorein malesthan

that in the females．In this connection．a recent study

reported that the AR gene was rarely expressed in the

poorly differentiated HCC affecting the males[55]．This

三


a

b

C

d Methylation％ C irrhosis(一)

C

N

C irrhosis(+)

C

N

T oral

C

N

Inhnnnnkn HI|L -．

- L|L l

n

C irrhosis (+)

■-I

T o tal

L I

LbL- _．_一_·-·一

=

善鏊呈董


●

一

一

Tab2．Summaryof bothOCCHITenC~and c(1ll‘mcyofthehypermethylationforeachgenesineachofHCCpatien t samples alongwith

s0lmeclinical-pathological parameters(age，gender,gradingandcirrhosis)．a)，Thenon~irrhosispatientgroup，b)，thecirrhosispatient

group，and c)，thetotalpatientgroup．Th efilled，shadingand emptyboxesindicatethe caseswhereonlyhypermethylated,both

hypermethylated and unmethylated and only unmethylated alleles were detected，respectively． Th e frequency(％)of the

hypermethylatedtargets(exceptforthe MAGEA1，wherethe hypomethylated)amongthetotalcaseswascalculatedand presented

inplot Th efrequencyofchanges(％)inthemethylationpatterninHCC(C)andtheneighboringnon-canceroustissues(N)againstthe normalhealthyliver tissuesweredetailedin d1．

does correlate well with the observation in this study

that there was high occurrence(82．14％，20／28)of the hypomethylation ofthe AR gene in the male HCC pa—

tienttissue．AlthoughitWas hypermethylated more fie—

quentlyinthefemale(83_3％．5／6)thanthemalepatient

group(68％，15／22)，the implication may be different． Loss of the hypermethylation of the AR gene in the

femalesma y symbolizethetumorassociateddefectsin

327

器

％％ ∞ ∞

++ ++ + + + + + ++ + + + +

9

C f m m m m f m m m mf m m m m 暮耋 ∞ ∞ 0 -罟 ∞ o l l

” ”蛇 “ " 砬．一戛薹一邑暑芎= 一富一Ilo!Ili 星

蚕。《苫量；曼量嚣量

一自基堇堇堇詈量詈

sIN 詈堇堇詈堇堇

，I皇堇詈詈詈堇堇

童墨量詈詈堇墨

一Ⅱ)n 堇堇堇詈堇堇

呈 v。堇量詈詈堇堇

量堇堇堇 g．。堇堇

>《u 8．。8．。 g．。暑．。 8．。8．。

，I目v 詈堇 g．。 g。堇 g．。

妄呈妻 g．。 g．。 g．。墨堇

《一皇 ∞f；l妻暮暮一圣三

妻譬妻 g．R毫室三

莹 ∞u ∞曼曼 g．。三．。量墨

置《0 。 69．

cf。 69．

．一 ∞冀一

f】 fl

一 g．。

[I

譬 g。

cf[f g。

[；g。

g．。

[f

雩一N !卜

∞ 是!

叶 !+

g 。 [f

冀c5一；一

冀c5 ∞

一8 堇要 g．。善要奏关

卺耄蚤墨善墓一蠡

爱耄；警嚣茎茎

§ 詈；薹 cff5 }rc8妻


M ethylationprofiling and concordantmethylationbehaviorsinHCC http：／／www．cell-research．tom

epigenetic control in genera1．

The DBCCRl gene identified at the region(q32一q33、 within—chromosome9qwith higllLOHinhumanbladder

carcinomaisthemostfrequentgeneticalterationin tran．

sitional cell carcinoma(TCC)of the bladder．Its loss of expression correlated well the hypermethylation of the

promoterCpGislandatafrequencyashighas 52％(36／ 69)bladder careinoma[50]．In this study，the promoter

CpG islan d has been detected at a comparable frequency

(53．57％．indicating its possible role in the HCC formation．

砀 e p73 gene encodes a protein structurally an d func．

tionallyhomologoustoTP53，an dmapsto chromosomal

ban d lp36-33．where loss of heterozygosity has been

observedin upto90％ ofoligodendrogliomas an din 10．

25％ 0f diffuse astrocytoma[56．571．In this report．we

foundthatthep73 genewasprevalentlymethylated(68 ％1 ill HCC tissues．However，whether its hypermethy．

1ation correlates with its functional inactivation remain

to be determined．

Both gen eticdefectsan depigeneticabnormalitiescon．

cemingthe、̂，T1 genehavebeenimplicatedinthefor．

marionof 妇 stumor[58] Inthis study,we alsofound

the W T 1 gene hypermethylated in 29％ of the HCC

eases，implyingitspossibleinvolvementin theformation

ofHCC．

Resistance of tumors to the cytotoxic chemothera．

pies may result from the disrupted apoptosis programs

and remains a major obstacle in cancer仃eatment． In

this connection，the IRF7 gene was also analyzed．the

analogue(IRF 1、of which has been implicated in"the ⅢN gammar mediatedapoptosiswitllaprofoundimpart

tothechemo—sensitivityofthetumorcells[59．601 功e

IRF7 expression was negatively regulated by the pro—

motermethylation[61]．Inthis study．the IRF7 gene was hypermethylatedin 46％ oftheHCCcasesbeing studied．

indicatingthepo ssible involvementofits inactivationby

thehypermethylationin HCC．

Hypomethylation ofthepromoter G islandofthe

M AGEA1 gene may be essential to its HCC spe—

cific expression Althoughthehypermethylationmediatedgene silenc—

ing of the tumor suppressor genes has caught the major attention,thelocalhypomethylationhasalsorecentlybeen

linked to reactivation of tran scription of the genes that

arehypermethylated an d silenced in the norm al tissues

328

『9，621．Therefore．we have also assessed whether the

swi tching．onoftheotherwisehypermethylatedgenesin

the norm al cells is linked to the demethylated state of

the promoter CpG island in the HCC tissues． The

MAGEA1 gene is such an example．expression ofwhich

is off in the norm al hepatocytes and on in HCC[3 11．

Correlating well with such an expression profile，we in

deed found that the promoter CpG island of the

MAGEA1(panel 13，Fig 1)wasfullymethylatedinthe norm al liver tissue，becam e unmethylated at a similar

frequency in 1 8／28(64％)and 2 1／28(75％)in each of

the paired HCC tissue sam ples，respectively．Judged by

the lack of the significance difference( =0．76，P=0．

382)．thedemethylation0fthepromoterCpGislandmay

more likely to be an early phase event of HCC

carcinogenesis．

The stage—specific nature ofthe HCC associated

changes in the methylation profiles ofthe promoter

CpG islands ofgenes It has been well recogn ized that the so．．called non．．

cancerouscellsdefmedun dermicroscopemayhavea1．

ready suffered some genetic lesions as have the corre．

sponding cancerous tissues，theoutcome of the earlier

events of carcinogenesis．Inclusion of the neighboring

non—can cerous tissue in this study made it possible to

analyzed our results from the stage．．specific perspec．．

tive ofcarcinogenesis．As shown in Tab 3．the follow．

ing genes exhibited a similar frequency in changes of

methylation paRern in the HCC tissues an d the neigh．

boring non—cancerous tissues from the paRern in the

norm allivertissuesfromhealthvolun teers：ABO(9／28．

32％；14／28，50％；)c =1．845，P=0．174)，CSPG2(1／28，

4％；5／28，1 8％，)Ck 2．987，P=0．084)，GALR2(2／28，

7％；6／28，21％；X2=2．292，P=0．13)，MT1A(2／28，7％，

3／28，11％；)C =0．2 1 6，P=0．642)，MYOD I(1 5／28，54％；

1 5／28，54％；P=1)，p57 (2／28，7％；4／28，14％；)C2=0．

707，P=0．388)，and WT1(4／28，14％；8／28，29％；)C =1．

697．P=0．193)as well as the MAGEA1 with having the HCC specific demethylation(18／28，64％；21／28，75％：

Z2=0．76，P=0．383)．Three of the four genes studied pre—

viously[27】displayed a similar pattern．They were：the

RASSFla(24／29，79％；29／29，100％；P=-0．085)，CDH13

(5／29，17％；6／29，20％，)CL 0．1 12，P=0．738)；and CASP8 (21／21，100％；21／21，100％：P=11．The rest fe11 into the

second category．Th ey are the genes encoding AR(12／ 28，43％；20／29，71％；X~--4．667，P=0．031)，cyclin al(8／



28，29％；15／28，54％；)C ：3．615，O．057)，DBCCR1(3／

28，1l％；16／28，57％；)C =13．462，p<0．001)，IRF7(2／28，

7％；12／28，43％；)C2==9．524，P=0．002)，OCT6(14／28，5O％；

23／28，82％；)C =6．452，P=0．01 1)，and p73(10／28，36％；

19／28，68％；)C 5．793，P=0．016)．The corresponding

figures ofthe pl6 ain the previously study『27]were

6／29，2O％，and 16／29，55％(X =7．323，P=0，007)．Ifthe neighboring non．．can cerous tissue may indeed have a1．．

ready suffered fxom the early stage genetic and／or epi—

genetic lesions during carcinogenesis，the HCC specific

methylation changes of the ABO，CSPG2，GALR2，

MT1A，MYOD1，W T1，Ⅳ【AGEA1，RASSF1a，CDH13

and CASP8 genes occurred at the early stage of the

HCC carcinogenesis． On the contrary．the changes in

the methylation pattern ofthe AR，cyclin al，DBCCR1，

IRF7，OCT6，p73，and pl 61NK4a genes was likely to be

the late phase event Of HCC carcinogenesis．The impli—

cmion of this new analysis in the HCC remains to be

exploredinthefuture．

Association studies of the methylation profiles of

thegeneswith theclinical-pathologicalparameters

ofHCC

After completing the information-gathering ofthe me—

thylationprofilesoffortyfourgenes(twentyfour genes

in this report and tw enty genes in the previous report

【271)it naturally follows to identify the association be— tw een the HCC changes in the methylation ofeach tar-

get、 th an y given clinical—pathological parameters．By

a stringent statistic evaluation with both)C and P tests，

we looked for the association betw een the HCC spe—

cific methylation chan ges of each of the nineteen genes

(including four genes in the previous study，1)and vari—

OUS clinical—pathological parameters．Probably due to

the relatively smaller size of patient samples，few asso—

ciation survived fxom such a scrutiny．

HCCis classifiedintotwomajor subgroups，thoseas— sociated with cirrhosis and those without，which differ

fxom both pathological processes as well as the etiologi—

cal profiles[63，64]． Cirrhosis is a significant pre—HCC

pathologiclesionandcanbeeasilydiagnosedbythenon- invasive ultrasound m ethod． In this connection，the

hyperm ethylated cyclin a l gene occurred m ore fre—

quently in the non—cirrhosis HCC group(10／13，76．9％)

than in the cirrhosis HCC group(5／15，20％，)C ：3．615，

P=0．057)．Onthecontrary．thehypermethylatedp16INK4

gene was more prevalent in cirrhosis HCC patients(12／

16，75％)than the non-cirrhosis HCC patients(4／13，3O．

3％，)C =7．323，P=0．007)． Although the underlying mechan isms whereby such a cirrhosis based differential

methylationprofilehas beenbroughtabout，it’svaluefor

the prognostic evaluation ofthe clinical treatment ofthe

HCC patients should not be overlooked．Furthermore．

it is an ticipated that more associations betw een the me—

thylation of any given targets an d the clinica1．．pathologi．．

cal indicators will be discovered when more HCC pa-

tients are subjected to such a study in future．

The concordant methylation behavior ofthepro—

moter G islands ofthe genes in HCC As far as the number ofthe genes is concerned，our

a

b

The early stage—specific change in methylation

c／M

l4·l 4，28 1．845

21／21 董 6 23／29 0 ll2

5·23／28 2 987

6 22／28 2 292

2l·7／2 8 0 760

3 25／28 0 2l6

15／28鹫徽 §

29-0／29

—

Th— — — —

e late stage-spe—

c—

ifi— — —

c— ， — —

c．

h— —

an ges in methylatio—

n—

Tab 3，Th e stage—specific changes in the methylation profiles of

genes during carcinogenesis．By stringent statistic analyses(c2and

P value)．the difference in occurrences of DNA methylation

changes between the neighboring normal tissues(N)and the

HCC tissues(C)has been evaluated． a1，lists the genes where

no significant difference were detected,indicating these chan ges

in the methylation profile an early stage—specific event，whereas

the b1listthegeneswiththe significantdifferenceswere detected,

with a close association with the late stage 0f HCC formation．

Incolumn 2an d 3，a—b／creferstothenumberofthemethylated

— unnlethylated cases／the total cases(Except for the MAGElA

gene，where the order of a—b is reversed)．NM，refers to the methylated cases in the neighboring non—cancerous tissue；an d

CM the methylated cases in the HCC tissues．N．B—Due to the

particular feature in the RASSF1A．where it is lOO％ hyper-

methylated in HCC tissue．it is beyond the capacity of such

statistic an alyses．The intuitive decision was supported by the

s~rle an alysis with assum ed l／29 sample were not hyoerme—

thylated in HCC．

329

Ⅲ咖m mⅢ ～脚 n n O O O O●O O O

a R 7 b ̈

C ) f


MethylationprofilingandconcordantmethylationbehaviorsinHCC

present studies are likely to be most comprehensive in

theHCCfield,too111"bestknowledge．which enableus

to lookinto the concordantmethylationbehaviorofthe

nineteen genes in the HCC group and in various sub．

groupsforthefirsttime．Both occurrenceandfrequen—

cies of the HCC specific chan ges in various subsets of

two upward to nine genes among the nineteen genes

(includingfourgenes inprevious study[27])withthebest

sub-set selection method[29]．Although，CASP8 was

hypermethylatedin 21／29 samplestested，the remaining

eight samplesfailedtobeinformative，we excludedthe

CASP8gen efrom thisanalysis．As shownin It，5．the

frequencyofthetwogene(thehyperm ethylatedOCT6

and RASSFla1 was 82％ and ofthree gene(the form er

twoplus thehypermethylatedp73)was68％andoffour

gene subsets(the form er three plus the demethylated

MAGEA1)was 54％ inthe HCC patient group．Inthe

a Ntmaberofgenes

2

3

4

5

6

7

8

9

O~~ureilce

C Numberofgenes O~~ureilce

吲 set

fF。Ⅱ恤rgct set ,Groups 。

C irrho~ s

C lrrhosis

330

0ceureenen

0ccurll~JllCe

Frequency(％1

Tab 4．The cirrhosis·associated hypermethylation of the genes in

HCC．By stringent statisticanalyses~2andPvalue)，thefrequency- differencesin changesin DNA methylationofthegen esbetweenthe

cirrhosis an d non-cirrhosis patient groups have been an alyzed and

presented．

cirrhosis-HCC groups，the frequency ofthe two genes

(the hyperm ethylated RASSF l a and OCT6)was 87％．

and three gene sub—sets(the form er two genes plus the

p l 6 or p73)was 73％．The corresponding paaem ofthe non—cirrhosis patient group was different，the fre．

quencyofthetwogenes(thehypermethylatedRASSFla plus the hypomethylated MAGEAl or the hyperm e—

thylated 0CT6 or the hyperm ethylated cyclin a l 1 was

77％andofthethree gene sub—sets(thehypermethylated

The genes in sub-set

Frequency(％) Th e genes in sub—set

Frequency(％)

77 R A SsF 1

77 R A SSF 1

77 R A SSF 1

69 R A SSF 1a

54 R A SSF 1

54 RA SSF l

54 R A SSF 1

46 R A SSF l

46 R A SSF 1

46 R A SSF 1

Frequency(％)

Frequency(％1

Frequency(％)

Th egenesin sub-set

cyd in a1

Th e genesin sub—set

R A S sF1a o C T6

R A S SF1a o C T6

R A S SF1a M A G EA

R A S SF1a o C T6

R A S SFla cyclin a1

■—●—■●■—■■■● ■—■■■● ——●—■●■●■■■● ■●■■一 ●———■●■—■■■一■————■ cyclin a1 ———■●_ ■—■■■●

cyclin a1

cyclin a1

cyclin a1

cyelin a I

cyclin a1

Th e genes in sub-set

R A SSF I a O C T 6 073

R A SSF I a O C T 6 D 16⋯ ‘

RA SSF Ia O C T 6 073

R A SSF I a O C T 6 cyclin aI

1 he genes in sub-set

R A SSF 1a O C T 6 liftl6⋯

R A SSF la O C T 6 lift16⋯

R A SSF 1a o C T 6 M A G EA l

R A SSF la o C T 6 cyclin a1

R A S SF la o C T 6 cyclin al

■—■■■● ■—■■■● ■—■■■●

■■■_

●——● ●——● ●——● ●——●

Tab 5． The summary of the concordant methylation behavior of the hypermethylated genes．Base upon the best sub．set

selection method，the gene subsets of tw o to nine genes in the HCC or tw o to five genes in the cirrhosis an d non．cirrhosis

subgroupswerepresented．a)，theHCC as awhole；b)，Thenon·cirrhosispatientgroup，c)，the cirrhosispatientgroup；d-O，the

summaryofthetw o，threean dfourgene subsets，respectively．

童重lll

蒹

誉

器器器

llllll

鍪一

器

譬

篓∞

一


Cell Research 13(5)，Oct，2003 JianYU，HongYuZHANG，etal

RASSF la plus hypermethylated OCT6 and cyclin a1) was 69％．

SUM M ARIES

Tumor associated changes in the methylation profiles

of the promoter CpG islands of,chiefly，the suppressor

genes have been well docum ented，suggesting the pos—

sible role of the epigenetic m echanism s for gene

inactivation，as an alternative to the genetic lesions，in—

cluding deletion and mutation in tumors[7，65，661．

Therefore．methylation profiling would be a usefulinfor-

mation gathering process for a better understanding of

carcinogenesis as well as the better diagnostic，progn os—

tic an d even therapeutic m easures against tumors． In

this study．we expand further the list Of targets from

twenty[271 to forty four known genes for methylation profilingin thenormalhealthyliver,HCCtissuesan dthe

paired normal tissues，representing a most comprehen—

sive survey in the HCC．

Finally，the concordan t m ethylation profiles of these

nineteen genes summarized above(d—C 1’ab 5)may be

useful as the prognostic，possibly the diagnostic

biomarkers fo r HCC．They m ay serve the good epige—

netic m arkers to detect tulrlor cells from biopsies，serunl，

an d so forth，if both sensitivity and specificity of these

as says can be satisfactorily established． It may also be

useful to determine the methylation status ofthese mark-

ers in circulating tulrlor cells in blood or predict the sen—

sitivitytochemotherapy，andoveralltherapeuticoutcomes

ofth e HCC patients differing in cirrhosis status，grading

an d other clinical—pathological profiles．

Thanks are due to Jian Gup CH EN．Li Sheng

ZHANG．M eng Chap W U and Su Shen ZHEN for pro—

viding the patient’s samples，Jian Ren GU for his sup—

port at the early stage ofthis project and Jian Xin and Jian yin Guo fo r their assistance in the statistic treatment

of the data．This work was supported by the National

High Technology Research and Development Program

of China (863 Program) (200 1 AA2 1 70 1 1，

2002AA2Z3 3 52)，the Major State Basic Research De—

velopment Program of China r973 Program)

(G l 99805 l 0o4)，and the Science Foundation of Shang—

hai Municipal Government(02DJ14056)to JingDe ZHU．

REFERENCES

l Ferlay J．F Bray．P Pisani and D M Parkin．GOBOCAB 2000

Can cer Incidence．Mortality an d Prevalence W_0rldwide．．Lyon：

IARC Scienctific Publications．IARC Press．200 l

2 Parkin D．P Pisani an d J Ferlay．Global can cer statistics．CA

Can cer J．Clin．．1999；49：33—64．

3 Cancer Incidence and Mortality in China．1993一l997(Selected

Cities and Counties)．Beiiing：China Publishing House ofMedi—

cal Sciences an d Technologies．1 998

4 FeitelsonM A，B Sun ，N L SatirogluTufan ，JLiu，JPan an dZ

Lian ．Genetic mechanisms ofhepatocarcinogenesis．Oncogene

2002；21(16)：2593—604．

5 Kim J W and X W 、Ⅳang．Gene expression profiling of

preneoplastic liver disease an d liver can cer：a new era for im—

proved early detection and treatment ofthese deadly diseases?

Carcinogenesis 2003；24 )：363—9．

6 Wan gX W SPHussain．TIHuo，eta1．M olecular pathogenesis

ofhuman hepatocellular carcinoma．Toxicology 2002；181—182：

43—7．

7 Baylin S and T H Bestor．Altered methylation patterns in

can cer cell genomes：cause or consequence?Can cer Cell 2002；

l(41：299—305．

8 Esteller M ．Relevan ce 0fDNA methylation in the man agement

ofcancer．LancetOncol 2003；4(61：35l一8．

9 Esteller M an d J G Herman．Cancer as an epigenetic disease：

DNA methylationan d chromatinalterationsin human tumours．

J Pathol 2002；196(1)：1—7． 10 Jones P A an d D Takai．The role 0fDNA methylation in main—

malian epigenetics．Science 200l：293(5532)：l068—70．

1l Jaenisch R and A Bird．Epigenetic regulation ofgene expression：

how the genome integrates intrinsic an d environmental signals．

Nat Genet 2003；33 Suppl：245—54．

12 Jones P A．Epigenetics in carcinogenesis an d cancer prevention．

Ann N Y Acad Sci 2003；983：213—9．

13 Geiman T M and K D Robertson．Chromatin remodeling，his—

tone modifications．an d DNA methylation—how does it all fit

together?J Cell Biochem 2002；87(21：ll7—25．

14 Feninberg A．Can cer epigenetics takes center stage．Proc．Nat1．

Acad．Sci．U S．A．200 l：98(2)：392—4．

1 5 Eden A，F Gaudet，A Waghmare an d R Jaenisch．Chromosomal

instability and tumors promoted by DNA hypomethylation．

Science 2003；300(56181：455．

16 Gaudet F．J G Hodgson，A Eden，et a1．Induction oftumors in

mice by genomic hypomethylation．Science 2003；300(5618)： 489—92．

17 Chen R Z，U Pettersson，C Beard，L Jackson—Grusby and R

Jaenisch．DNA hypomethylation leads to elevated mutation

rates．Nature 1998；395(66971：89—93． 1 8 Esteller M，M F Fraga，M Guo，et a1．DNA methylation pat—

ternsin hereditaryhumancancersmimic sporadictumorigenesis．

Hum Mol Genet 2001：10(26)：3001—7．

19 Zochbauer—Muller S，K M Fong，A K Virm an i，J Geradts．A F

Gazdar and J D M inna．Aberran t promoter methylation of

multiple genes in non—small cell lun g can cers．Can cer Res 200l：

61(1)：249—55． 20Rosas SL，W Koch，M G daCosta Carvalho，et a1．Promoter

hyperm ethylation patterns ofp16．06一methylguan ine—DNA—

methyltransferase．an d death—associated protein kinase in tu一

331


McthylationprofilingandconcordantmethylationbehaviorsinHCC http：llwww．cell-research．corn

咖璐 and salivaofheadandneckcancerpailents．Cancer Res

20ol：610)：939-42． 21FosterS八 DJWong,M TBarrettan dD A Galloway．Inacti．

vation ofpl6 in humanmammary cpithelial cells by CpG

islandmethylation．Mol CenBi01 l998；18‘41：1793．801．

22CllirLPOnyango，SBrandenburg，YW_u，CLHsiehandAP

Feinberg．Loss of imprinting in colorectal can cer linked to

m hyla on ofH19 and IGF2．Cancer Res 2oo2；62(22)：

6442-6，

23 ChoB，H Lee9S Jeong，et a1．Promoterhypomethylationofa

novel cancer／testisan tigengen eCAGEis correlatedwithits

almnmltexpressionandis seenin pren mlignantstageofgastric

carcinoma．Biochcm BiophysResCommtm2003：307(1)：52．

63．

24 Hattori M，H Sakamoto。K Satoh and T Yamamoto．DNA

demethylas~isexpressedin ovarian cancersan dthe expression

correlates with demcthylation of CpG sites in the promoter

regionofc-erbB．2 and survivingenes．Can cer Lett2o0l：169

‘2l：l55 ．

25HomeS F’LABryce。GBWisman，eta1．Lackoftelomerase

RNA gen eh，I1ERC expressionin alternativelengthen ingofte—

lomeres cells is associated with methylation of the h1]巳RC

promoter．CancerRes2ool：61(1)：27．32．

26ToyotaM，NAhuja,M Ohe-Toyota，JGHerlnan。SBBaylin

an dJPIssa．CpO islan dmcthylatorphen otypein colorectal

cancer ProcNatlAcadSciUSA l999：9 l53：8681．6．

27 YII J，M Ni，J Xu，et a1．M ethylation profiling of twenty

promoter-CpO islandsofgeneswhichmaycontributetohcpa．

tocdhlarcarcinogenesis．BMCCancer2002：2(1)：29．

28 Liver Can cer Study Group of Japan ．TNM classification for

helmtoeellularcarcinomabyLiver Can cer Studygroup．w0rld

JStag l989；13：212．

29 M iller A J．Subset Selection in Regression．New York

Champman andHall l99o．

30 LjuQ，Wang,w &~ds．LiverCancer．Beijing：People’sMedi— calPublishingHouse．2oo0

3l MouDC，SLCai,JRPen g，eta1．EvaluationofMAGE．1 an d

M AGE-3 as tumour-specific markers to detect blood dissemi．

nationofhepatocellularcarcinoma cells．BrJCancer2002；86

(1)：ll0-6．

32 Y’m H an d K Blan chard．DNA methylation represses the ex．

pression of the human erythopoietin gene by two diflfferent

mechanisms．Blood 20o0；95(1)：ll 1．9．

33 Sauter P an d P Matthias．Coactivator OBF．1 makes selective

contacts with both the POU．specific domain and the PoU

homeod omain an d acts as a molecular clamp on DNA．M ol

Cen Bi0l l998：18(12)：7397-409．

341wamoto S，D A W ithers，K Handaan dSHak omori．Deletion

ofA．an tigen in a hum an can cer celllineis associatedwith

l'l~ lCedpromoter activityofCBF／NF-Y binding region，an d

possibly with enhanced DNA methylation of A transferase

promoter．Glycoconj J l 999；16(10)：659．66． 35Chen B，PDias，J J Jenkins，3rd,V H SavellandD M Parham．

M ethylation alterations of the MyoDl upstream region are

predictive ofsube lassification ofhuman rhabdomyosarcoma s．

Am J Pathol l998；152(4)：l071．9．

36HutchinsA S，A C Mullen，H W Lee，et a1．Gene silencing

quantitatively controlsthe functionofadevelopmentaltrans．

activator．MolCen2oo2：1O(1)：81．91．

332

37Reeben M ，SMyohanen ，M Saarmaan dH Prydz．Sequencing

of the rat light neurofilamen t promoter reveals differences in

methylation between expressing an d non-expressing cell lines，

but not tissues．Gene l995；157(1．2)：325．9．

38KibdA S．M Christopher,D AF柚．G SBova,PJGoodfellow

an d W B Isaacs．M ethylation an d mutational an alysis of p27

(kip1)inprostate carcinoma ．Prostate 2ool；48(4)：248．53．

39 Li Y．H Nagai，T Ohno，et a1．Aberrant DNA methylation of

p57(KIP2)geneinthepromoter regioninlymphoidma lignan-

ties of B．cell phenotype．Blood 2002：100(7)：2572-7． 40 Watanabe T．H Huang，M Nakamura,eta1．Methylati0nofthe

p73 gene in gliomas．Acta Neuropathol(Bed)2002；104(4)：

357．62．

41 LauxD E，EM Curran，W V W e1shons．D B Lubalm an dTH

Huang．Hypenneth~ lation of the W ilms’tunlor suppressor

gene CpG islan d in hum an breastcarcinomas．BreastCan cer

Res Treat l999；56(1)：35．43．

42WangS，THashcmi，SFiled．ALClemmonsandBEHawes．

Differential intracellular signaling of the GalRl an d GalR2

galanin receptor subtypes．Biochemistry 1998；37(19)：671 l-

7．

43 Usmani B A，R Shen，M Jan eczko，et a1．M ethylation of the

neutralen dopeptidasegen epromoterinhumanprostatecancers．

Clin Cancer Res 20oo；6f ：l664．70．

44Venkataraman G M ．M Yalin，R Maleinekan dK BAin．Res．

torationofiod ideuptakein dedifferentiatod thyroidcarcinoma ：

relationship to human Na+，I．symporter gene methylation

status．J Clin Endocrinol Metab l 999；84(7)：244 9．57．

45Cui J，LRRohr,G Swanson，V O Speights。TMaxwellandA

R Brothman．Hypermethylation of the caveolin．1 gene pro．

rooter in prostate cancer．Prostate 2ool：46(3)：249．56．

46ToyotaM，CHo。NAhuja,eta1．Identificationofdifferentially

methylated sequencesin colorectal cancer bymethylatedCpG

island am plification．Cancer Res l999；5 10)：2307．12．

47 Sasaki M，Y Tanaka，G Pednchery,et a1．Methylation an d

inactivationofestrogen ，progesterone，an dandrogen receptors

in prostate cancer．J Natl Cancer Inst 2002；94(5)：384．90． 48 M uller C，C Readhead，S Diederichs，et a1．M cthylation ofthe

cyclin Al promoter correlates with gene silen cing in soma tic

cell lines．while tissue．specific expression ofcyclin Al is me．

thylation independent．Mol Cell Biol 2000；20(9)：33 16．29． 49 Rachmilewitz E A．The role of methylation in CM L．Przegl

Lek 2000；57 Suppl 1：25．6．

50Habuchi M Luscombe ．PA Elderan dM A Knowles．Struc．

ture and methylation-based silencing of a gene(DBCCRl、 within a can didate bladder cancer tilmor suppressor region at

9q32．q33．Genomics l 998；48(3)：277．88．

5l Ghoshal K，S Majum der,Z Li，X Dong and S T Jacob．Sup．

pression ofmetallothionein gene expression in a rat hepatoma

because ofpromoter-specific DNA methylati0n．J Biol Chem

2000；275f1)：539．47．

52 Y_吣 a Y DNA me a n in cancer and ageing．M ech Ageing

Dev 2002：l23fl2)：l 649．54．

53黜chardSon B．Impact 0f ag_mg on DNA memylati0n．Age_mg I s I v 2003：2(3)：245．6 1．

54 Yamanaka M，M W taⅡabe，Y Yamad瓦 et a1． t部ed metlIvla．

0n Of multiple genes in carciⅡogenesis Of the prostate．Int J

Cancer 2003：l06(3)：382．7．

55 TavianD，G DePe吨，A Pit0zzi。N P0rt0lani。SM GiuliIliand


Cell Research 13(5)，Oct，2003 JianYU，HongYuZHANG，etal

S Bad ．Androgen receDtor mRNA under．expression in poorly

differentiated human heDatocellular carcinoma．Histol

Histopathol 2002；17(4)：ll l 3-9．

56DongS，JCPang，JHu，LFZhouan dH KNg．Transcriptional

inactivation ofTP73 expression in oligodendroglial tumors．Int

J Cancer 2002；98(3)：370—5．

57 Loiseau H．J Arsaut an d J Demotes．Mainard．p73 gene tran．

scripts in human brain tumors-overexpression an d altered splic．

g in ependymomas．Neurosci Lett l 999；263(2—3)：l 73．6．

58 Satoh Y，T Nakagawachi，H Nakadate，et a1．Significant Reduc．

tion of W T l Gene Expression．Possibly Due to Epigenetic

AIteration in Wilms’Tumor．J Biochem (Tokyo)2003：133

(3)：303．8． 59To mitaY VBilim，N Ham，TKasaharaandK Talkahashi．Role

Of II F．1 an d caspase．7 in IFN．gamma enhancement of Fas．

mediated apoptosis in ACHN renal cell carcinoma cells．Int J

Cancer 2003；104(4)：400．8．

60DetjenKM，DMurphy,M Welzel，KFarwig，BWiedenmann and S Rosewicz．Downregulation of p2 l(waffcip．1)mediates

apoptosis ofhum an hepatocellular careinoma cells in response

to interferon．gamma．Exp Cell Res 2003；282(2)：78-89． 6l LuR，W CAu，W SYeow,N Hageman an dPM Pitha．Regu·

lation of the promoter activity of interferon regulatory factor-

7 gene．Activation by interferon snd silencing by hyperme-

thylation．J Biol Chem 2000；275(41)：3 l 805．1 2．

62 Ellrlich M ．DNA methylation in can cer：too much．but also too

little．Oncogene 2002；21(351：5400．13．

63 Su Q and P Bannasch．Relevance of hepatic preneoplasia for

hum an hepatocarcinogenesis．Toxicol Pathol 2003；31(1)：l26-

33．

64 Kopp．Schneider A．Biostatistical evaluation of focal hepatic

preneoplasia．To xicol Pathol 2003；31(1)：l21．5． 65 Baylin S B and J G Herman．DNA hypermethylation in

tumorigenesis：epigenetics joins genetics．Trends Genet 2000； 16(4)：l 68．74．

66 Bestor T H．Unan swered questions about the role of promoter

methylation in carcinogenesis．Ann N Y Acad Sci 2003；983：

22．7．

333


Documents

Methylation profiling of twenty four genes and the concordant methylation behaviours of nineteen genes that may contribute to hepatocellular carcinogenesis