Early during Myelomagenesis Alterations in DNA Methylation That Occur Myeloma Is ...

153 Pages·2013·2.47 MB·English

Myeloma Is Characterized by Stage-Specific
Alterations in DNA Methylation That Occur
Early during Myelomagenesis
Christoph J. Heuck, Jayesh Mehta, Tushar Bhagat, Krishna
This information is current as
Gundabolu, Yiting Yu, Shahper Khan, Grigoris Chrysofakis,
of March 3, 2019.
Carolina Schinke, Joseph Tariman, Eric Vickrey, Natalie
Pulliam, Sangeeta Nischal, Li Zhou, Sanchari Bhattacharya,
Richard Meagher, Caroline Hu, Shahina Maqbool, Masako
Suzuki, Samir Parekh, Frederic Reu, Ulrich Steidl, John
Greally, Amit Verma and Seema B. Singhal
DDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDD
J Immunol published online 13 February 2013 oooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooo
wwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwww
http://www.jimmunol.org/content/early/2013/02/13/jimmun nlonlonlonlonlonlonlonlonlonlonlonlonlonlonlonlonlonlonlonlonlonlonlonlonlonlonlonlonlonlonlonlonlonlonlonlonlonlonlonlonlonlonlonlonlonlonlonlonlonlonlonlonlonlonlonlonlonlonlonlonlonlonlonlonlonlonlonlonlonlonlonlonlonlonlonlonlonlonlonlonlonlonlonlonlonlonlonlonlonlonlonlonlonlonlonlonlonlonlonlonlonlonlonlonlonlonlonlonlonlonlonlonlonlonlonlonlonlonlonlonlonlonlonlonlonlonlonlonlonlonlonlonlonlonlonlonlonlonlonlonlonlonlonlo
ol.1202493 adadadadadadadadadadadadadadadadadadadadadadadadadadadadadadadadadadadadadadadadadadadadadadadadadadadadadadadadadadadadadadadadadadadadadadadadadadadadadadadadadadadadadadadadadadadadadadadadadadadadadadadadadadadadadadadadadadadadadadadadadadadadadadadadadadadadadadadadadadadadadadadad
eeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeee
dddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddd
fro fro fro fro fro fro fro fro fro fro fro fro fro fro fro fro fro fro fro fro fro fro fro fro fro fro fro fro fro fro fro fro fro fro fro fro fro fro fro fro fro fro fro fro fro fro fro fro fro fro fro fro fro fro fro fro fro fro fro fro fro fro fro fro fro fro fro fro fro fro fro fro fro fro fro fro fro fro fro fro fro fro fro fro fro fro fro fro fro fro fro fro fro fro fro fro fro fro fro fro fro fro fro fro fro fro fro fro fro fro fro fro fro fro fro fro fro fro fro fro fro fro fro fro fro fro fro fro fro fro fro fro fro fro fro fro fro fro fro fro fro fro fro fro
mmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmm
h h h h h h h h h h h h h h h h h h h h h h h h h h h h h h h h h h h h h h h h h h h h h h h h h h h h h h h h h h h h h h h h h h h h h h h h h h h h h h h h h h h h h h h h h h h h h h h h h h h h h h h h h h h h h h h h h h h h h h h h h h h h h h h h h h h h h h h h h h h h h h h h
SuppleMmaetnetraiarly 3h.tDtpC:/1/www.jimmunol.org/content/suppl/2013/02/13/jimmunol.120249 ttp://wttp://wttp://wttp://wttp://wttp://wttp://wttp://wttp://wttp://wttp://wttp://wttp://wttp://wttp://wttp://wttp://wttp://wttp://wttp://wttp://wttp://wttp://wttp://wttp://wttp://wttp://wttp://wttp://wttp://wttp://wttp://wttp://wttp://wttp://wttp://wttp://wttp://wttp://wttp://wttp://wttp://wttp://wttp://wttp://wttp://wttp://wttp://wttp://wttp://wttp://wttp://wttp://wttp://wttp://wttp://wttp://wttp://wttp://wttp://wttp://wttp://wttp://wttp://wttp://wttp://wttp://wttp://wttp://wttp://wttp://wttp://wttp://wttp://wttp://wttp://wttp://wttp://wttp://wttp://wttp://wttp://wttp://wttp://wttp://wttp://wttp://wttp://wttp://wttp://wttp://wttp://wttp://wttp://wttp://wttp://wttp://wttp://wttp://wttp://wttp://wttp://wttp://wttp://wttp://wttp://wttp://wttp://wttp://wttp://wttp://wttp://wttp://wttp://wttp://wttp://wttp://wttp://wttp://wttp://wttp://wttp://wttp://wttp://wttp://wttp://wttp://wttp://wttp://wttp://wttp://wttp://wttp://wttp://wttp://wttp://wttp://wttp://wttp://wttp://wttp://wttp://wttp://wttp://w
wwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwww
wwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwww
.jim.jim.jim.jim.jim.jim.jim.jim.jim.jim.jim.jim.jim.jim.jim.jim.jim.jim.jim.jim.jim.jim.jim.jim.jim.jim.jim.jim.jim.jim.jim.jim.jim.jim.jim.jim.jim.jim.jim.jim.jim.jim.jim.jim.jim.jim.jim.jim.jim.jim.jim.jim.jim.jim.jim.jim.jim.jim.jim.jim.jim.jim.jim.jim.jim.jim.jim.jim.jim.jim.jim.jim.jim.jim.jim.jim.jim.jim.jim.jim.jim.jim.jim.jim.jim.jim.jim.jim.jim.jim.jim.jim.jim.jim.jim.jim.jim.jim.jim.jim.jim.jim.jim.jim.jim.jim.jim.jim.jim.jim.jim.jim.jim.jim.jim.jim.jim.jim.jim.jim.jim.jim.jim.jim.jim.jim.jim.jim.jim.jim.jim.jim.jim.jim.jim.jim.jim.jim.jim.jim.jim.jim.jim.jim
mmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmm
Why The JI? Submit online. uuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuu
nnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnn
• Rapid Reviews! 30 days* from submission to initial decision ol.orgol.orgol.orgol.orgol.orgol.orgol.orgol.orgol.orgol.orgol.orgol.orgol.orgol.orgol.orgol.orgol.orgol.orgol.orgol.orgol.orgol.orgol.orgol.orgol.orgol.orgol.orgol.orgol.orgol.orgol.orgol.orgol.orgol.orgol.orgol.orgol.orgol.orgol.orgol.orgol.orgol.orgol.orgol.orgol.orgol.orgol.orgol.orgol.orgol.orgol.orgol.orgol.orgol.orgol.orgol.orgol.orgol.orgol.orgol.orgol.orgol.orgol.orgol.orgol.orgol.orgol.orgol.orgol.orgol.orgol.orgol.orgol.orgol.orgol.orgol.orgol.orgol.orgol.orgol.orgol.orgol.orgol.orgol.orgol.orgol.orgol.orgol.orgol.orgol.orgol.orgol.orgol.orgol.orgol.orgol.orgol.orgol.orgol.orgol.orgol.orgol.orgol.orgol.orgol.orgol.orgol.orgol.orgol.orgol.orgol.orgol.orgol.orgol.orgol.orgol.orgol.orgol.orgol.orgol.orgol.orgol.orgol.orgol.orgol.orgol.orgol.orgol.orgol.orgol.orgol.orgol.orgol.orgol.orgol.orgol.orgol.orgol.orgol.orgol.orgol.orgol.orgol.orgol.org
b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ b/
yyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyy
• No Triage! Every submission reviewed by practicing scientists g g g g g g g g g g g g g g g g g g g g g g g g g g g g g g g g g g g g g g g g g g g g g g g g g g g g g g g g g g g g g g g g g g g g g g g g g g g g g g g g g g g g g g g g g g g g g g g g g g g g g g g g g g g g g g g g g g g g g g g g g g g g g g g g g g g g g g g g g g g g g g g g
uuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuu
eeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeee
ssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssss
• Fast Publication! 4 weeks from acceptance to publication t ot ot ot ot ot ot ot ot ot ot ot ot ot ot ot ot ot ot ot ot ot ot ot ot ot ot ot ot ot ot ot ot ot ot ot ot ot ot ot ot ot ot ot ot ot ot ot ot ot ot ot ot ot ot ot ot ot ot ot ot ot ot ot ot ot ot ot ot ot ot ot ot ot ot ot ot ot ot ot ot ot ot ot ot ot ot ot ot ot ot ot ot ot ot ot ot ot ot ot ot ot ot ot ot ot ot ot ot ot ot ot ot ot ot ot ot ot ot ot ot ot ot ot ot ot ot ot ot ot ot ot ot ot ot ot ot ot ot ot ot ot ot ot o
nnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnn
M M M M M M M M M M M M M M M M M M M M M M M M M M M M M M M M M M M M M M M M M M M M M M M M M M M M M M M M M M M M M M M M M M M M M M M M M M M M M M M M M M M M M M M M M M M M M M M M M M M M M M M M M M M M M M M M M M M M M M M M M M M M M M M M M M M M M M M M M M M M M M M M
*average arcarcarcarcarcarcarcarcarcarcarcarcarcarcarcarcarcarcarcarcarcarcarcarcarcarcarcarcarcarcarcarcarcarcarcarcarcarcarcarcarcarcarcarcarcarcarcarcarcarcarcarcarcarcarcarcarcarcarcarcarcarcarcarcarcarcarcarcarcarcarcarcarcarcarcarcarcarcarcarcarcarcarcarcarcarcarcarcarcarcarcarcarcarcarcarcarcarcarcarcarcarcarcarcarcarcarcarcarcarcarcarcarcarcarcarcarcarcarcarcarcarcarcarcarcarcarcarcarcarcarcarcarcarcarcarcarcarcarcarcarcarcarcarc
hhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhh
3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3
Subscription Information about subscribing to The Journal of Immunology is online at: , 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2
000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000
http://jimmunol.org/subscription 191919191919191919191919191919191919191919191919191919191919191919191919191919191919191919191919191919191919191919191919191919191919191919191919191919191919191919191919191919191919191919191919191919191919191919191919191919191919191919191919191919191919191919191919191919191919191919191919
Permissions Submit copyright permission requests at:
http://www.aai.org/About/Publications/JI/copyright.html
Email Alerts Receive free email-alerts when new articles cite this article. Sign up at:
http://jimmunol.org/alerts
The Journal of Immunology is published twice each month by
The American Association of Immunologists, Inc.,
1451 Rockville Pike, Suite 650, Rockville, MD 20852
Copyright © 2013 by The American Association of
Immunologists, Inc. All rights reserved.
Print ISSN: 0022-1767 Online ISSN: 1550-6606.
Published February 13, 2013, doi:10.4049/jimmunol.1202493
The JournalofImmunology
Myeloma Is Characterized by Stage-Specific Alterations in
DNA Methylation That Occur Early during Myelomagenesis
Christoph J. Heuck,* Jayesh Mehta,† Tushar Bhagat,* Krishna Gundabolu,*
Yiting Yu,* Shahper Khan,‡ Grigoris Chrysofakis,* Carolina Schinke,* Joseph Tariman,†
Eric Vickrey,† Natalie Pulliam,† Sangeeta Nischal,* Li Zhou,* Sanchari Bhattacharya,*
Richard Meagher,† Caroline Hu,* Shahina Maqbool,* Masako Suzuki,* Samir Parekh,*
Frederic Reu,‡ Ulrich Steidl,* John Greally,* Amit Verma,* and Seema B. Singhal†
Epigeneticchangesplayimportantrolesincarcinogenesisandinfluenceinitialstepsinneoplastictransformationbyalteringge-
nomestabilityandregulatinggeneexpression.Tocharacterizeepigenomicchangesduringthetransformationofnormalplasma
cellstomyeloma,wemodifiedtheHpaIItinyfragmentenrichmentbyligation–mediatedPCRassaytoworkwithsmallnumbersof
purifiedprimarymarrowplasmacells.Thenano-HpaIItinyfragmentenrichmentbyligation–mediatedPCRassaywasusedto
D
analyze the methylome of CD138+ cells from 56 subjects representing premalignant (monoclonal gammopathy of uncertain o
w
significance),early,andadvancedstagesofmyeloma,aswellashealthycontrols.Plasmacellsfrompremalignantandearlystages nlo
ofmyelomawerecharacterizedbystriking,widespreadhypomethylation.Gene-specifichypermethylationwasseentooccurinthe ad
e
advancedstages,andcelllinesrepresentativeofrelapsedcaseswerefoundtobesensitivetodecitabine.Aberrantdemethylationin d
monoclonal gammopathyof uncertain significance occurred primarily in CpG islands, whereas differentially methylated loci in fro
m
cases of myeloma occurred predominantly outside of CpG islands and affected distinct sets of gene pathways, demonstrating h
qualitative epigenetic differences between premalignant and malignant stages. Examination of the methylation machinery ttp
revealed that the methyltransferase, DNMT3A, was aberrantly hypermethylated and underexpressed, but not mutated in mye- ://w
w
loma.DNMT3Aunderexpressionwasalsoassociatedwithadverseoverallsurvivalinalargecohortofpatients,providinginsights w
into genesis of hypomethylation in myeloma. These results demonstrate widespread, stage-specific epigenetic changes during .jim
myelomagenesis and suggest that early demethylation can be a potential contributor to genome instability seen in myeloma. m
u
We also identify DNMT3A expression as a novel prognostic biomarker and suggest that relapsed cases can be therapeutically n
o
targeted byhypomethylatingagents. TheJournalofImmunology,2013, 190: 000–000. l.o
rg
b/
D espitetherapeuticadvances,multiplemyeloma(MM)re- thedevelopmentofariskmodelthatremainsrobustevenintheage y g
mains incurable and needs newer insights into the path- ofnewertherapies(1).Theuseofgeneexpressionprofiling by ue
s
ogenicmechanismsthatcausethisdisease.Geneexpres- different groups has also led to the identification of distinct mo- t o
sionprofilinghasbeenusedextensivelyinmyelomaandhasledto lecular subgroups that show defined clinical features (2, 3). How- n M
ever, pathogenic mechanisms driving these differences in gene a
rc
expressionhavenotbeenwelldescribed,especiallyforearlystages h
3
*AlbertEinsteinCollegeofMedicine,Bronx,NY10461;†NorthwesternUniversity, inmyelomagenesis(4). , 2
FeinbergSchoolofMedicine,Chicago,IL60611;and‡ClevelandClinic,Cleveland, Recentstudiesoftheepigenomeandinparticularthemethylome 01
OH44195 haveshownthatcancerischaracterizedbywidespreadepigenetic 9
ReceivedforpublicationSeptember7,2012.Accepted forpublicationJanuary9, changes. These changes lead to altered gene expression that can
2013.
result in activation of oncogenic pathways. Furthermore, methyl-
ThisworkwassupportedbyNationalInstitutesofHealthImmunooncologyTraining ome profiling has shown greater prognostic ability than gene ex-
Program GrantT32 CA009173,Gabrielle’s Angel Foundation,the Leukemiaand
LymphomaSociety,theAmericanCancerSociety,theDepartmentofDefense,and pressionprofilinginacutemyeloidleukemia(AML)andhasledto
aPartnershipforCuresgrant. identification of newer molecular subgroups with differences in
ThemethylationdatapresentedinthisarticlehavebeendepositedintheNational overall survival (5). We have shown that methylome profiling is
InstitutesofHealthGeneExpressionOmnibus(NationalCenterforBiotechnology abletoidentifychangesthatoccurearlyduringcarcinogenesisin
Information,http://www.ncbi.nlm.nih.gov/geo/)underaccessionnumberGSE43860.
solid tumors such as esophageal cancer (6). These studies have
AddresscorrespondenceandreprintrequeststoDr.AmitVerma,Dr.ChristophJ.
been conducted with the use of the HpaII tiny fragment enrich-
Heuck,andDr.SeemaB.Singhal,AlbertEinsteinCollegeofMedicine,Bronx,NY
10461(A.V.andC.J.H.)andFeinbergSchoolofMedicine,NorthwesternUniversity, ment by ligation–mediated PCR (HELP) assay, which is a ge-
Chicago,IL60611(S.B.S.).E-mailaddresses:[email protected](A.V.), nome-wide assay that provides a reproducible analysis of the
[email protected](C.J.H.),[email protected](S.B.S.)
methylome that is not biased toward CpG islands (7). Analyzing
Theonlineversionofthisarticlecontainssupplementalmaterial.
the methylome of myeloma can help identify changes that can
Abbreviationsusedinthisarticle:AML,acutemyeloidleukemia;HAF,HpaII-amplifi-
define disease subsets and lead to identification of newer thera-
ablefragment;HELP,HpaIItinyfragmentenrichmentbyligation–mediatedPCR;MM,
multiplemyeloma;MGUS,monoclonalgammopathyofuncertainsignificance; peutic targets. Recent sequencing studies in myeloma have also
NEWMM,newlydiagnosedmyeloma;NL,normal;REL,relapsedmyeloma;REM, revealed mutations in enzymes that are involved in epigenetic
completeremission;SMM,smolderingmyeloma;TF,transcriptionfactor.
machinery, again reinforcing the need to study the epigenetic
Copyright(cid:1)2013byTheAmericanAssociationofImmunologists,Inc.0022-1767/13/$16.00 alterationsinthisdisease(8).Wehaveconductedagenome-wide
www.jimmunol.org/cgi/doi/10.4049/jimmunol.1202493
2 METHYLOME OFMULTIPLEMYELOMA
analysis of changes in DNA methylation in monoclonal gamm- tativeformethylationandanalyzedasacontinuousvariable.Formostloci,
opathyofuncertainsignificance(MGUS),newlydiagnosedMM, each fragment was categorized as either methylated, if the centered log
HpaII/MspIratiowasgenerally,0,orhypomethylated,if,incontrast,the
aswellasrelapsedMM,andcomparedthemwithnormalplasma
logratiowas.0.
cell controls. We used a modification of the HELP assay (nano-
HELP)thatwasdevelopedtoworkwithlowamountsofDNAto QuantitativeDNAmethylation analysisbyMassArray
interrogate the methylome of sorted CD138+ cells from patient Epityping
samples. We report that widespread alterations in DNA methyla-
ValidationofthefindingsoftheHELPassaywascarriedoutbyMALDI-
tion are seen in myeloma and have the power to discriminate TOFmassspectrometryusingEpiTyperbyMassArray(Sequenom,CA)on
between MGUS and new/relapsed cases. We report that hypo- bisulfite-convertedDNA,aspreviouslydescribed(6).Validationwasdone
methylation is the predominant early change during myeloma- onallsampleswithsufficientavailableDNA.MassARRAYprimerswere
designedtocovertheflankingHpaIIsitesforagivenHAF,aswellasfor
genesis that is gradually transformed to hypermethylation in
anyotherHpaIIsitesfoundupto2000bpupstreamofthedownstreamsite
relapsed cases, thus providing the epigenetic basis for the differ- andupto2000bpdownstreamoftheupstreamsite,tocoverallpossible
entiation betweenthese differentstages of myeloma. alternativesitesofdigestionwithinourrangeofPCRamplification.
Microarraydataanalysis
Materials and Methods
Patientsamples Unsupervised clustering of HELP data by hierarchical clustering was
performedusingthestatisticalsoftwareRversion2.6.2.Atwo-samplettest
Bonemarrowaspirateswereobtainedwithsignedinformedconsentfrom was used for each gene to summarize methylation differences between
11 patients with MGUS, 4 patients with smoldering myeloma (SMM), groups.Geneswererankedonthebasisofthisteststatistic,andasetoftop
13patientswithnewlydiagnosedmyeloma(NEWMM),16patientswith differentially methylated genes with an observed log fold change of .1
relapsedmyeloma(REL),including2patientswithserialsamples,and2 between group means was identified. Genes were further grouped
D
patients in clinical complete remission (REM) who are followed at the accordingtothedirectionof themethylationchange (hypomethylated o
w
myeloma clinic at the Robert H. Lurie Cancer Center at Northwestern versus hypermethylated), and the relative frequencies of these changes n
University. All specimens underwent CD138+ microbead selection (Mil- were computed among the top candidates to explore global methylation lo
a
tenyiBiotec).PuritywasconfirmedbyflowcytometrystainingforCD38 patterns.ValidationwithMassArrayshowedgoodcorrelationwiththedata d
andCD45.Sampleswithapurity,90%byflowcytometrywerenotused generated by the HELP assay. MassArray analysis validated significant ed
for this study.Eight samplesfrom normaldonorswithoutknown malig- quantitativedifferencesinmethylationfordifferentiallymethylatedgenes fro
nancieswereobtainedcommerciallyfromAllCells(Emeryville,CA).All selectedbyourapproach. m
normaldonorsampleshadapurityof$85%. h
Sequencing ofDNMT3A ttp
DNAmethylation analysisusingthenano-HELPassay ScreeningformutationsintheDNMT3Acatalyticdomainwascarriedout ://w
GenomicDNAwasextractedusingtheGentraPureGenekitusing13105 usingdirectgenomicsequencing.PCRprimersweredesignedtoamplify ww
atoss5ay,3as10d5esccerilblse.dDpNreAviomuestlhyy(l7a,ti9o)n.analysis was done using the HELP a2n0dnsgeqgueennocmeiccoDdiNngAewxoanssu1s8e–d2f3or(aPvCaiRlabalmepulpiofincarteioqnu,esfto)l.lFoworeedacbhyPpCurRi-, .jim
m
Inbrief,100nggenomicDNA(nano-HELPmodification)wasdigested ficationusingMontageCleanupkit(Millipore,Billerica,MA).Sequencing
u
overnight using the two isoschizomers HpaII and MspI. The digestion was performed using ABI 3730xl DNA analyzer (Applied Biosystems, n
o
productswerethenpurifiedoncewithphenol-chloroform.Afterovernight FosterCity,CA).PCRwascarriedoutusingthefollowing:94˚C,4min, l.o
adapterligation,theHpaIIandMspIproductswereamplifiedinatwo-step followed by 35 cycles of 94˚C, 30 s, specific annealing temperature for rg
protocol,aspreviouslydescribed(7).HpaII-andMspI-generatedgenomic eachprimer,30s,andextension72˚C,30s,final72˚C,5min.Resultswere b/
fragments between 200 and 2000 bp in length were labeled with either comparedwithreference(NM_175629)andSNPdatabases(dbSNP;http:// y g
Cy3-orCy5-labeledrandomprimersandthencohybridizedontoahuman www.ncbi.nlm.nih.gov/projects/SNP).Ifchromatogramssuggestedapos- u
e
(HG17)custom-designedoligonucleotidearray(50mers)covering25,626 siblemutation,bidirectionalresequencingwasperformed. s
HpaII-amplifiablefragments(HAFs)annotatedto14,214genepromoters. t o
HAFsaredefinedasgenomicsequencescontainedbetweentwoflanking Pathwayanalysisandtranscriptionfactorbindingsiteanalysis n M
HpaII sites found 200–2,000 bp apart. Each HAF on the array is repre- a
sented by a probe set consisting of 14–15 individual probes, randomly TheIngenuityPathwayAnalysissoftware(IPA,RedwoodCity,CA)was rc
used to determine biological pathways associated with differentially h
distributedacrossthemicroarrayslide.Methylationdatapresentedinthis 3
articlehavebeendepositedintheNationalInstitutesofHealthGeneEx- methylatedgenes,as performedpreviously(6,10).Enrichmentofgenes , 2
associatedwithspecificcanonicalpathwayswasdeterminedrelativetothe 0
pressionOmnibus(NationalCenterforBiotechnologyInformation,http:// 1
Ingenuityknowledgedatabaseforeachoftheindividualplatformsandthe 9
www.ncbi.nlm.nih.gov/geo/)underaccessionnumberGSE43860. integratedanalysisatasignificancelevelofp,0.01.Biologicalnetworks
HELP dataanalysisandqualitycontrol capturedbythemicroarrayplatformsweregeneratedusingIPAandscored
basedontherelationshipbetweenthetotalnumberofgenesinthespecific
Allmicroarrayhybridizationsweresubjectedtoextensivequalitycontrol networkandthetotalnumberofgenesidentifiedbythemicroarrayanal-
using the following strategies. First, uniformity of hybridization was ysis. The list of differentially methylated genes was examined for en-
evaluatedusingamodifiedversionofapreviouslypublishedalgorithm(10, richment of conserved gene-associated transcription factor (TF) binding
11) adapted for the NimbleGen platform, and any hybridization with sitesusingIPAaswellasotherpublishedgenesetsavailablethroughthe
strongregionalartifactswasdiscardedandrepeated.Second,normalized MolecularSignaturesDatabase(MSigDB)(12).
signalintensitiesfromeacharraywerecomparedagainsta20%trimmed
mean of signal intensities across all arrays in that experiment, and any Meta-analysisofmyelomageneexpressionstudies
arraysdisplayingasignificantintensitybiasthatcouldnotbeexplainedby
WeobtainedgeneexpressiondatafromtheArkansasdatasetsGSE5900and
thebiologyofthesamplewereexcluded.SignalintensitiesateachHpaII-
GSE2658(4)fromNationalCenterforBiotechnologyInformation’sGene
amplifiablefragmentwerecalculatedasarobust(25%trimmed)meanof
ExpressionOmnibus database.Thedatasetswere quantilenormalizedto
theircomponentprobe-levelsignalintensities.Anyfragmentsfoundwithin
ensurecross-studycomparability,basedonourpreviousapproach(13,14).
the level of background MspI signal intensity, measured as 2.5 mean-
AnalyseswereperformedusingSAS(SASInstitute,Cary,NC)andtheR
absolute-differences above the median of random probe signals, were
language (http://www.r-project.org). The final database had 22 normal
categorizedasfailed.Thesefailedlocithereforerepresentthepopulation
controls,44MGUS,and559newMMsamples.
offragmentsthatdidnotamplifybyPCR,whateverthebiological(e.g.,
genomic deletions and other sequence errors) or experimental cause. In Cell linesandcultureconditions
contrast, methylated loci were so designated when the level of HpaII
signal intensity was similarly indistinguishable from background. PCR- Cell lines U266 and RPMI8226 were purchased from American Type
amplifying fragments (those not flagged as either methylated or failed) CultureCollection(Manassas,VA).CelllinesMM1.SandMM1.Rwere
were normalized using an intra-array quantile approach wherein HpaII/ providedbyS.Rosen(NorthwesternUniversity,Chicago,IL).Allcelllines
MspI ratios are aligned across density-dependent sliding windows of wereculturedinRPMI1640medium(Invitrogen,Carlsbad,CA)supple-
fragmentsize-sorteddata.Thelog (HpaII/MspI)wasusedasarepresen- mentedwith10%heat-inactivatedFBSandpenicillin(100U/ml),strep-
2
TheJournal ofImmunology 3
tomycin(100mg/ml),and4mMglutamine.Cellsweremaintainedat37˚C markers in tumor models (5). To determine whether there was
andhumidifiedwith95%airand5%CO2forcellculture.Fortheviability aberrant differential methylation in different stages of multiple
assays, the cells were cultured in 0.5, 1, and 5 mM decitabine (Sigma- myeloma,weusedthenano-HELPassay(7)onCD138+selected
Aldrich)for5d.Decitabinewasaddedtotheculturedaily;DMSOwas
bonemarrowcellsfrom11patientswithmonoclonalgammopathy
servedascontrol.Viabilitywasmeasuredonday6usingMTS(Promega),
accordingtothemanufacturer’sinstruction. of uncertain significance (MGUS), 4 patients with SMM, 13
patients with NEWMM, and 16 patients with REL, which also
included 2 patients with serial samples. We also included 2 my-
Results
eloma patients in REM. CD138+ selected bone marrow plasma
Genome-wide analysisofDNAmethylation revealsepigenetic
cells from 8 healthy donors (normal [NL]) served as controls.
alterations inplasmacellsfrompatientswithmyelomaand
Clinical characteristics of the 48 patients with plasma cell dys-
MGUS
crasias are listed in SupplementalTable 1.
Extensivestudyofgeneexpressionprofilingofmyelomacellshas Atfirstweperformedanunsupervisedanalysisofthegenerated
led to several molecular models that allow the classification of methylation profiles with hierarchical clustering and the nearest
myelomaindifferentriskcategories,whichhasledtosignificant shrunkencentroidalgorithm.Bothmethodsshowedthatthehealthy
improvements in treatment strategies and outcomes (1, 15). Epi- controlsformedatightclusterthatwasdistinctfromsampleswith
genetic alterations including aberrant DNA methylation can reg- abnormal plasma cells, demonstrating epigenetic dissimilarity
ulate gene expression and can also be used as better prognostic betweenthesegroups(Fig.1).Interestingly,MGUSsamplesalso
D
o
w
n
lo
a
d
e
d
fro
m
h
ttp
://w
w
w
.jim
m
u
n
o
l.o
rg
b/
y
g
u
e
s
t o
n
M
a
rc
h
3
, 2
0
1
9
FIGURE1. DNAmethylationpatternscandifferentiatebetweendifferentstagesofplasmacelldyscrasias.(A)Unsupervisedhierarchicalclusteringof
usingmethylationprofilesgeneratedbytheHELPassayseparatesthesamplesintwomajorclusterscontainingthemajorityofNEWMM(orange)andREL
(red)samplesorthemajorityofNL(green)andMGUS(lightblue)samples,respectively.Thesetwoclusterswerealsoidentifiedbythepresence(dark
green)orabsence(darkblue)ofabnormalitiesdetectedbyFISHorconventionalcytogenetics.Theclustersareeachfurtherseparatedintotwosubgroups,
resultinginatotaloffourcohortsrepresentingthemajorityofNL,MGUS,NEWMM,andREL.ThebottomsixlinesrepresentFISHdata,green=not
detectedbyFISH,pink=detectedbyFISH,gray=notdone.♦ordindicatespairedsamples.(B)Unsupervisedthree-dimensionalclusteringbasedon
nearest shrunken centroid algorithm using methylation profiles also shows distinction between normal and myeloma samples. Among the myeloma
samples,clusteringofMGUSsamplesisdistinctfromnewandrelapsedcases.Green,NL;blue,MGUS;yellow,SMM;orange,NEWMM;red,REL;
purple,REM.
4 METHYLOME OFMULTIPLEMYELOMA
clustered in a distinct group, suggesting definite alterations in inNEWMMsamples,yetthiswaslesspronouncedthaninMGUS
DNA methylation that occur early during myelomagenesis. The (Fig. 2B, Supplemental Table 2C, 2D). Interestingly, in cases of
new (NEW) and relapsed (REL) myeloma cases demonstrated relapsed myeloma, we observed increased hypermethylation,
great epigenetic dissimilarity to the NL and MGUS samples and demonstrating that there was a trend toward predominant hypo-
separatedintwosubgroupswiththemajorityofNEWMMinone methylation in MGUS to predominant hypermethylation in REL
and the majority of REL samples in the other group (Fig. 1A). (Fig. 2C, 2D, Supplemental Table 2E, 2F). Overall, differential
These findings show that methylation profiling can differentiate methylatedgenesatthetransitionbetweenstagesaredepictedin
between NL, MGUS, NEWMM, and REL, in an unsupervised Supplemental Fig. 1A. When analyzing this differential methyl-
manner. Included in our analysis were two sets of consecutive ationatthetransitionfromonediseasestatetothenext(i.e.,NLto
samplestakenfromthesamepatientatdifferenttimepoints.Inthe MGUS, MGUS to NEWMM, and NEWMM to REL), we found
first case, the initial sample (REL15) clusters with the group of that of the 2963 unique genes that were hypomethylated at the
relapsedmyelomaandtheposttreatmentsampleclusterswiththe transition from NL to MGUS, 2472 became hypermethylated at
NEWMM samples (REL9). The second case included samples the transition of MGUS to NEWMM. Of those, 2 were also sig-
takenatdistincttimepointswithoutanytreatmentsintheinterval. nificantly hypermethylated at the transition from NEWMM to
Both samples cluster together in the REL group (REL18, REL6) REL, albeit with a q value .0.05 (p , 0.05, Supplemental
and thusshowthebiological validityof ouranalysis. Fig.1B).Alistofthedifferentiallymethylatedgenesbetweenthe
differentdiseasestatescanbefoundinSupplementalTable2A–K.
MGUS andNEWMMshowpredominanthypomethylation,
The HELP assay has been validated quantitatively in many
whereasRELarepredominantlyhypermethylated
studies(5,6),andwealsovalidatedourfindingsbyanalyzingthe
After demonstrating global epigenetic dissimilarity between nor- methylation status of differentially methylated genes, ARID4B D
mal and myelomatous plasma cells, we wanted to determine the and DNMT3A, by bisulfite Massarray analysis. Examination of o
w
specific differences in DNA methylation between the different the promoter regions of these genes demonstrated a strong cor- n
lo
stagesofmyeloma.Weperformedasupervisedanalysiscomparing relation of quantitative methylation values obtained from Mas- a
d
the MGUS, NEWMM, or REL cases versus control samples. sArraywiththefindingsofHELPassay,demonstratingthevalidity ed
Volcanoplotscomparingthedifferenceofmeanmethylationofall ofourfindings(SupplementalFig.2). fro
individuallociwereplottedagainstthesignificance(log[pvalue] m
Differentiallymethylatedgenesshowspecificfunctional and h
abparpisseivndaglouanent,atbe0sso.t0)lu0ot5efwftohelerdedcuihfsafeenrdgeentocoefid$(eFn2igti(.f=y2l)od.giSf[ftHerirpneagnIetIin/aMtllcsypuItm-]oe.ftfhs1yc)laoatmnedd- gWeenonmeixctcahnaarlayczteedristhtiecsbiological pathways that were associated ttp://ww
loci. All probes thus identified were significant after multiple withdifferentiallymethylatedgenesinmyelomaandobservedthat w
testingwithBenjamini-Hochbergcorrectionwithafalsediscovery pathwaysinvolvedincellproliferation,geneexpression,cellcycle, .jim
rate of ,5%. We observed that majority of differentially meth- and cancer were significantly involved by aberrantly methylated m
u
ylated loci in MGUS were found to be hypomethylated when genesinMGUS,NEWMM,aswellasrelapsedcasesofMM(Table no
comparedwithnormalplasmacells(Fig.2A,SupplementalTable I).Thegenesaffectedbyaberrantmethylationofthesepathways l.o
rg
2A,2B).Hypomethylationwasthepredominantchangealsoseen includedmanygenesthatwerehypomethylated(CEBPalpha,ILs, b/
y
g
u
e
s
t o
n
M
a
rc
h
3
, 2
0
1
9
FIGURE2. MGUSandNEWMMshowpredom-
inant hypomethylation, whereas REL are predomi-
nantly hypermethylated. Volcano plots showing
difference of mean methylation (x-axis) and signifi-
canceofthedifference(y-axis)demonstrateaberrant
hypomethylation in MGUS (A) and both hypo- and
hypermethylationinnew(B)andrelapsed(C)cases
ofmyeloma.Thenumberofdifferentiallymethylated
HAFsisindicatedaboveeachvolcanoplot.Numbers
totheleftindicatehypomethylatedHAFs,andnum-
bers to the right indicate hypermethylated HAFs.
Aberrant hypermethylation is the predominant
changeinrelapsedcases(D).
TheJournal ofImmunology 5
TableI. Genesandpathwaysaberrantlymethylatedinmyeloma
BiologicalFunctions Genes
GenesHypomethylatedinMGUS
1 Cellmorphology,cellularassembly ADNP,AFF2,ARL15,B4GALT3,BRSK2,CASC3,DIS3L,
andorganization,cellularfunction EBAG9,EXOSC1,EXOSC2,EXOSC8,FBXO9,FLNC,GDI2,
andmaintenance IL15,MRPS15,NAGA,NGF,NR3C2,P2RX3,PLCG1,PRKCZ,
PRPH,STARD10,STK11,TGFA,TNK1,TSPAN7,TUBE1,
UPF3B,WDR6,ZNF74,ZNF83
2 Cellsignaling,moleculartransport, FFAR3,FZD1,FZD5,FZD10,GALR1,GPR6,GPR26,GPR44,
nucleicacidmetabolism GPR68,GPR77,GPR84,GPR87,GPR132,GPR137,GPR149,
GPR153,GPR161,GPR176,GPRASP2,GRM6,HRH2,LGR5,
LPHN1,MC3R,NPY2R,NPY5R,OPRK1,PTGER1,PTGER3,
QRFPR,SCTR,TAS1R2,VIPR1
3 Cellulardevelopment, ALOX5,ALX4,C12orf44,CEBPE,CLEC11A,COPB2,COPE,
hematologicalsystemdevelopment COPG,COPG2,COPZ1,CXCL5,DDA1,ETS2,FARSB,FLI1,
andfunction,hematopoiesis GFI1,GPX1,HMGB2,HSD17B4,IL6R,NAA15,NAA16,
S100A9,SACM1L,SIPA1L3,SPI1,SPIB,SRGN,TARS,TDP2,
UBA5,UQCRC2,ZXDC
4 Cellcycle,hairandskin BANF1,BLOC1S1,CCNE1,CHEK1,CHMP5,CHMP2A,
developmentandfunction, CUL4A,CUL4B,DCAF11,DCAF16,DDB1,ERCC8,GRP,
embryonicdevelopment KAT2A,LIN28B,MED20,NEK9,NUDCD1,PHIP,PKM2,
PRPF31,SART3,TADA1,TBL3,USP4,USP5,USP8,USP35,
D
USP36,USP37,USP43,WDR5B,ZNF277 o
5 Aminoacidmetabolism,cellular ANKRD28,AP1G1,ATP2A3,CCDC47,CCT7,CLN3,CLPX, w
n
assemblyandorganization,genetic DDIT4L,DGKZ,DPM1,GAR1,GEMIN4,GEMIN5,HOOK1, lo
disorder LOC100290142/USMG5,NECAP1,NHP2,NUP93,PFDN2, ad
e
SEC61A1,SH2D3C,SHC1,SLC25A10,SLC25A11,SLC25A22, d
SMAP1,SNRPG,TNFRSF14,TSC22D1,UNC45A,VBP1,WDR8 fro
GenesHypermethylatedinRelapsedMM m
1 Geneticdisorder,inflammatory AMH,ATF5,BATF,BTG2,CHPF,CHSY3,CTH,EXOSC1, h
disease,respiratorydisease EXOSC3,EXOSC6,EXOSC10,FERMT3,GDF9,GET4,ILF3, ttp
LMO2,LSM3,LSM7,LTBP4,MAGED1,NAA38,NOD1, ://w
NOD2,NUDT21,PBK,RIPK2,SCLT1,SCRIB,SGTA, w
SNAPC5,SUGT1,TRIB3,UNC5A w
2 Geneexpression,cellular ACY1,AKR1B1,BHLHE41,CSMD1,DLX4,ERG,EZH2, .jim
development,cellulargrowthand FRZB,GNB4,GNB5,HEXIM2,HOXA11,HOXB2,HOXB4, m
proliferation IER5L,KRT6A,LHX2,NPM3,RFTN1,RGS9,SLC29A1, u
n
SOX2,SOX4,SOX14,SYTL1,TNF,TNKS2,TNKS1BP1, o
TTC1,ZBTB11 l.o
3 Cellsignaling,moleculartransport, ADRB1,BAI3,CCR6,CNR2,CXCR1,CYSLTR1,GAB1, rg
vitaminandmineralmetabolism GABBR2,GNG7,GPER,GPR1,GPR15,GPR25,GPR44, b/
y
GPR62,GPR88,GPR125,GPR146,GPR151,GPRC5A,LGR4, g
u
LPAR2,MC4R,NPY,PIK3CG,PTGDR,PTGER1,RXFP4,SSTR2 e
s
4 Geneticdisorder,neurological APEX1,CDCA8,CDK16,CDK19,CDK5R1,CORO1C, t o
disease,dermatologicaldiseases COX7A2L,EIF6,EIF4B,GDI1,GDI2,GNL1,HOXC13,LZTS1, n
andconditions MSN,NOL3,NUDT5,OSGEP,PAICS,PLEK,RAB5C,RCC2, M
a
RPS6,RUFY1,SAP30BP,SCT,SET,TPM2,TUFM,ZNF212 rc
5 DNAreplication,recombination APLF,ARHGEF4,DLX2,DNTT,ERCC4,GATA6,HEYL, h
3
andrepair,cellmorphology, IKZF1,KIF5A,LIG4,MECOM,MSX2,PCNA,PDCD6, , 2
cellularfunctionandmaintenance PECAM1,POLI,RFC4,RPA1,SP8,TAOK3,VPS28,VPS37B, 0
1
WNT5B,WRN,XPA,XRCC2,XRCC6 9
GenesHypomethylatedinNewMM
1 Cellsignaling,carbohydrate ADORA1,BDKRB1,BDKRB2,CHRM2,FFAR1,FFAR3,
metabolism,lipidmetabolism FPR1,GABBR2,GLP2R,GPR20,GPR32,GPR113,GPR120,
GPR139,GPR161,GPR109B,GPRASP2,HBEGF,MC2R,
MCHR1,MTNR1A,NMUR1,NPY5R,OPN1LW,OXGR1,
RGR,RHO,TAC3,TACR2,TAS1R2,VN1R4
2 Neurologicaldisease, CAPN9,CAPN11,CDH4,CDH7,CDH17,CST3,DNAH1,
immunologicaldisease, DNAH5,DNAH10,DNAH17,DNAI2,EDN3,EID1,EIF4EBP2,
inflammatorydisease IL1R1,IL1RAPL1,LIG3,MAPK6,MBP,MYOZ1,NRAP,
PDLIM2,PLP1,RNASE2,SIGIRR,STK39,TGM2
3 Agpresentation,cell-to-cell BGN,DOCK1,EFNA3,Hsp70,IFNA1/IFNA13,IgG,IL1,IL25,
signalingandinteraction, IL12,IL17A,JPH3,KIR2DL3,KRT2,KRT3,KRT9,KRT14,
hematologicalsystemdevelopment KRT20,KRT23,LBP,LGALS7/LGALS7B,MYLK2,POTEKP,
andfunction S100A3,SELP,SERPINB3,SERPINB4,SH3GL2,Tlr,
TNFRSF1B,TREM1,TUBA3C/TUBA3D
4 Cancer,connectivetissuedisorders, ACR,ACTG2,ADCY,ADCY8,cacn,CACNA1B,Cacng,
reproductivesystemdevelopment CACNG1,CACNG3,CACNG5,CNGA3,CNR1,COTL1,
andfunction GNAO1,GNG2,KIF23,LPO,MYLK3,MYO7A,PCP4,PDE6G,
POU2F2,PPP3R2,PSCA,RIT2,SHROOM3,SPTB,TUBB8,VIL1
5 Lipidmetabolism,molecular ABCG5,ABCG8,ADAMTS13,Akt,AMPK,APOB,CD22,
transport,smallmolecule CEBPA,COL15A1,COL5A3,COL6A3,CYP3A4,CYP4F2,
biochemistry CYP8B1,FCAR,GRB10,HNF4adimer,KLK8,LIPE,MRC2,
NR0B2,NR1H4,PCK1,PTPN3,SCARB1,SLC22A7,STAR,
T3-TR-RXR,VWF
(Tablecontinues)
6 METHYLOME OFMULTIPLEMYELOMA
TableI. (Continued)
BiologicalFunctions Genes
GenesHypermethylatedinNewMM
1 Geneexpression,aminoacid ACTR3,ANP32E,BTAF1,CBX5,CDK9,DPYSL2,ERCC4,
metabolism,posttranslational FUBP1,GTF2H1,HNRNPA0,HNRNPA1,HNRNPUL1,
modification HOXA11,LSM7,MDFIC,MEPCE,NAA38,NNT,PLOD3,
POU2AF1,PRMT5,PRUNE,SART3,TAF4,TAF11,TUBB
2 Geneticdisorder,inflammatory AATF,AIMP1,AIMP2,BARD1,EFNA1,EPHA5,FXC1,
disease,respiratorydisease HNRNPC,HSF1,LMO3,LZTR1,NEDD9,NOD1,NOD2,
NR1H3,SIRT1,SS18L1,STMN2,SUGT1,TIMM9,TIMM8A,
TOMM40L,TOMM70A,TRIP6,XRCC6,ZNF584
3 DNAreplication,recombination, ACACB,AKT1S1,ARFIP2,ATF5,BTG2,CD34,CDKN2A,
andrepair,cellcycle,cellular CDKN2B,CENPK,CHAF1A,CHTF18,GSC,
development HIST1H2AB/HIST1H2AE,ID3,PCNA,PLEKHA8,POLD1,
RFC4,RPS16,SLC3A2,TBX5,TCF19,UBR7,ULK1,WRN
4 Gastrointestinaldisease,organismal ANKS1A,APOB,BRMS1,C14orf156,DNAJA1,DNAJB11,
injuryandabnormalities,genetic DNAJC9,EML1,EVX1,FOS,FTL,KCNQ4,KIAA1279,LCP1,
disorder PDPK1,PIK3CA,PRKCB,RASSF5,SAP18,SET,SGK3,
SLC4A10,SPTBN2,TPM2
5 Lymphoidtissuestructureand BIRC2,BPGM,C1orf56,CCDC71,CHAT,DHX15,FAM81A,
development,tissuemorphology, GGA3,ID2,KCNC4,NDUFS3,NFKBIE,NOLC1,PECI,POGK,
moleculartransport POLK,PTEN,PTPN2,SLC9A3R1,SLITRK1,SOX4,STC1,
D
TRAPPC3,ZNF264/ZNF805 o
w
n
lo
a
d
and their receptors, GFI1, etc.) and hypermethylated (EZH2, significantlyfoundtooccurwithinCpGislands,whereasbothnew e
d
variousHOXmembers,SOX,andWNTfamilymembers)thathave andrelapsedcasesofMMhadaberrantlymethylatedlocilocated fro
not beenpreviouslyimplicated inmyelomagenesis. outside of CpG islands (Fig. 3). This shows that there are quali- m
h
Aberrant methylation was not distributed randomly across tative differences in epigenetic alterations between MGUS and ttp
chromosomes.DifferentiallymethylatedHpaIIfragmentsshowed myeloma and is also consistent with recent findings that have ://w
significant regional differences with positional association with highlightedthataberrantmethylationincancercanoccuroutside w
chromosomes 1, 9, 11, 15, 17, 19, 20, and 21. Furthermore, to of CpGislands (16). w
determinewhethertheseaberrantlymethylatedregionssharedany .jim
common DNA elements, we performed a search for TF binding Genes involved inDNAmethylation machineryare m
sites enriched in these loci. Significant overrepresentation of differentiallymethylatedandexpressedinmyeloma uno
binding sites for TFs that have been implicated (PAX, MEF, and BecausewesawanoveralldecreaseinmethylationinMGUS,we l.o
MAZ) in myelomagenesis was observed (Table II). Various TFs analyzedourdatasetforthemethylationstatusofgenesinvolvedin brg/
suchasHNF6,PAX4,STAT3,EVI1,andothersweresignificantly thisprocessandalsoevaluatedfortheexpressionofthesegenesin y
g
enriched in relapsed cases of MM, including some that have not large gene expression datasets (Arkansas datasets GSE5900 and u
e
been implicated inthepathophysiologyof MM previously. GSE2658)thathavebeenpublishedpreviously(4).Specifically,we st o
Finally,wealsosoughttodeterminewhetherCpGislandswere analyzedthemethyltransferasesDNMT1,DNMT3a,andDNMT3b n
M
predominantly affected by differential methylation in myeloma. and observed that DNMT3a was significantly underexpressed in
a
We observed that aberrant methylation occurring in MGUS was a large independent cohort of myeloma gene expression profiles rc
h
3
, 2
0
1
TableII. Transcriptionfactorbindingsitesenrichedindifferentiallymethylatedloci 9
TranscriptionFactor GenesinOverlap Motif
MGUS
ARNT 202 NDDNNCACGTGNNNNN
ATF6 95 TGACGTGG
E47 205 VSNGCAGGTGKNCNN
USF 203 NNRNCACGTGNYNN
LXR 55 TGGGGTYACTNNCGGTCA
TCF1P 184 GKCRGKTT
NewMM
PAX2 36 NNNNGTCANGNRTKANNNN
GATA 161 WGATARN
MEF2 21 NNTGTTACTAAAAATAGAAMNN
E2F 56 TWSGCGCGAAAAYKR
CACCCbindingfactor 205 CANCCNNWGGGTGDGG
RelapsedMM
STAT3 113 NNNTTCCN
PAX4 193 NAAWAATTANS
EGR3 56 NTGCGTGGGCGK
EVI1 44 AGATAAGATAA
HNF6 189 HWAAATCAATAW
MAZ 135 GGGGAGGG
SREBP1 136 NATCACGTGAY
TheJournal ofImmunology 7
The observation of global hypomethylation in MGUS builds
upontworecentstudiesthatnoticedlossofmethylationinplasma
cellneoplasms(18,19).ThefirststudybySalhiaetal.(19)reports
hypomethylation as early as in MGUS and noted no difference
between methylation of new and treated myeloma samples. The
second study by Walker et al. (18) noticed hypomethylation in
myeloma samples, butdid not seeanysignificantdifferences be-
tween normal plasma cells and MGUS samples. Remethylation
was seen in plasma cell leukemia samples in comparison with
MM. In contrast to the latter study, which interrogated 27,578
CpGsites(correspondingto14,495genes),theformerstudyused
an approach looking at a limited set of only 1,505 CpG sites
(corresponding to 807 genes), thus limiting generalizability of
those findings. The study presented in this work was able to dis-
tinguish between normal, MGUS, and myeloma (new and re-
lapsed)samplesonunsupervisedclusteringandrevealedclear-cut,
widespreaddifferencesbetweenthesegroups.WeusedtheHELP
FIGURE 3. Differential methylation in MGUS occurs mainly within assay, a high-resolution assay, that is not biased toward CpG is-
CpG islands. The genomic location of differentially methylated regions lands and has been able to show stage-specific differences in
wasmappedtoCpGislandsforeachsubcategoryofmyeloma.DMRsin various tumor models (5, 20). This assay analyzes 25,626 CpG D
MGUSweresignificantlyenrichedwithinCPGislands(darkgray)com- sites and thus is comparable to the system used by Walker et al. o
w
paredwiththepercentageofprobesinthewholearray(proportionstest, (18). Just like Walker et al. (18), we observe relative hyper- n
p,0.01).DMRsinNEWMMandRELweresignificantlylocatedoutside lo
ofCpGislands(proportionstest,p,0.01). methylation in late stages of MM; however, we observe hypo- ad
e
methylation much earlier, that is, already at the level of MGUS, d
andnotonlyMM,similartothereportsofSalhiaetal.(19).This fro
m
(Fig. 4A) and had a significant impact on survival. After sepa- discrepancymightbeduetothelimitednumberofNLandMGUS h
rating patients into four quartiles according to their DNMT3A samples (3 and 4, respectively) used by Walker et al. (18). Our ttp
expression, we found that high DNMT3A expression (Q4) at findingsofearlyhypomethylationinMGUS,whichismaintained ://w
baseline conveyed a significantly overall survival compared with throughouttheearlyMMstage,butthenconvertstopredominant w
w
pgarotiuepn,tslowgirtahnkvetreyst,lopw=D0N.0M1,TF3igA.4eCxp).reIsnsioounrc(noh=ort14o0fpiantieenactsh, hmyapgeernmeesitsh.ylation, thus represent a novel assessment of myelo- .jim
m
we found the promoter of DNMT3A to be aberrantly hyper- ThefindingofhypomethylationinMGUSandmyelomaisalso u
n
methylated (Fig. 4B). Mutations of the gene encoding DNMT3a differentfromtheprevioussinglelocusstudiesthathavefocused o
haverecentlybeenreportedformyelodysplasticsyndrome(MDS) onhypermethylationoftumorsuppressorgenesinmyeloma(21). l.org
and AML (17) and have not been examined in myeloma. Se- The classical cancer-associated epigenetic alteration is promoter b/
quencingofthecatalyticdomainofDNMT3Ain23ofourCD138 CpG island hypermethylation. Even though global hypo- y g
purifiedplasmacellsamplesrevealednoexonicmutationorSNPs. methylation was reported in the pioneering epigenetic studies in ue
s
These results suggest that DNMT3A expression in myeloma is cancer (22), most investigators have subsequently focused on t o
mainlyreduced byaberrant hypermethylation. hypermethylationinCpGislandswithinselectedgenepromoters. n M
Array-basedDNAmethylationassaysthathavemainlyfocusedon a
MyelomacelllinesaresensitivetoDNMTinhibitors rc
CpG islands have supported the biased perception that CpG is- h
Lastly,becauseweobservedthatrelapsedcasesofmyelomaexhibit landsareuniquelyresponsibleformethyl-DNA–inducedgenomic 3, 2
increased aberrant methylation, wewanted to determine the sen- changes. Newer iterations of assays for the analysis of DNA 0
1
sitivity of these cells to hypomethylating agents. We tested the methylation,whicharebasedonnext-generationsequencing,have 9
efficacyofDNMTinhibitor,decitabine,inlongstandingmyeloma revealed a much more complex picture with DNA methylation
cell lines that exhibit various cytogenetic alterations and are re- occurring further upstream in CpG island shores, in introns, as
flective of relapsed cases. Decitabine was able to significantly wellasextendingmuchfurtherdownstream.Withthisinmind,it
inhibittheproliferationofthesemyelomacelllinesatbothlowand is difficult to speculate about the true meaning of differential
high doses (Fig. 5). In fact, the dose of 0.5 mM has shown to be methylationinCpGislands,andthisclearlypointstotheneedof
hypomethylating without causing DNA damage and was signifi- further in-depth investigations with these newer methods. Hypo-
cantly able to inhibit myeloma cell proliferation, thus suggesting methylation has been hypothesized to lead to carcinogenesis by
that reversal of hypermethylation can be a potential therapeutic promoting genomic instability (23, 24) as well as by aberrant
strategyinthese cases. activation of oncogenes (23). Because myeloma is characterized
byvarious chromosomal translocations anddeletions, thefinding
Discussion of early hypomethylation may be an important pathogenic
Weshowthatmyelomaischaracterizedbywidespreadalterations mechanismthatpromotessecondarygeneticeventsthatleadtothe
in DNA methylation that are specific for different stages of the development offull-blowndisease.
disease. Early stage of MGUS is characterized by predominant Efforts to distinguish MGUS from MM in an unsupervised
hypomethylationthatisprevalentenoughtodistinguishthesecells manner using gene expression profiling data from large clinical
from normal control plasma cells. The later stages acquire pro- trialshavenotbeensuccessfultodate(25,26).AlthoughMGUS
gressive hypermethylation with maximum methylation seen in samples can readily be separated from NL samples, they appear
relapsed cases. These data provide a comprehensive epigenomic identical to MM at a gene expression profiling level. Using
map of myelomagenesis and also identify important oncogenic methylationdata,wewereabletoclearlydistinguishbetweenthe
gene pathwaysthat are targeted bythese aberrant changes. majorityofNL,MGUS,MM,andRELsamples.Geneexpression
8 METHYLOME OFMULTIPLEMYELOMA
D
o
w
n
lo
a
d
e
d
fro
m
h
ttp
://w
w
w
.jim
m
u
n
o
l.o
rg
b/
y
g
u
e
2F2I)G,UMRGEUS4.(nD=N44M),Ta3nAdMisMund(ner=ex5p5r9es)sferdomanAdrhkyapnesarmsdetahtayslaettesdGiSnEm59y0e0loamnad.G(AS)EB26o5x8pslohtoswshsoigwniinfigcagnetnlyereexdpurceesdsioenxpvraelsuseiosninleCveDls13in8+mcyeelllosmfraom(ttNesLt,(pn,= st on
0.05).TheMMsamplesincludedinthesedatasetscontaineduntreatedsamples.(B)BoxplotsrepresentingthemethylationoftheDNMT3Apromoterinnormal M
a
plasmacells(NL),MGUS,andmyelomasamples(MM)showhypermethylationinMMcomparedwithNL(ttest,p,0.05).(C)LowDNMT3Aexpressionis rc
h
associatedwithworseoverallsurvivalinTT2.DifferencesbetweengroupswithtopandbottomquartilegeneexpressionareshownwithKaplan–Meiergraphs. 3
, 2
0
profiling can be affected by large-scale changes in just a few ationofDNA,itisnotgloballybiasedbydifferencesinafewloci, 1
9
transcripts. Because methylation analysis is dependent on evalu- andthusisnotaffectedbychangesthatmayoccuronlyinami-
nority of the analyzed cells. It is therefore more reflective of bi-
ology of premalignant conditions, as illustrated by our previous
study in esophageal carcinogenesis (6). In the present study, we
used CD138+ selection to isolate the myelomatous and normal
plasma cells; thus, it is possible that the observed difference be-
tween MGUS and MM samples is in part due to dilution with
normalsamples.Strategiessuchasmultiparameterflowcytometry
sortingarebeingusedtoensureclonalityinfuturestudies.Atthis
pointitshouldbenotedthat,althoughtwosamplesincludedinour
study were in clinical remission with ,3% of plasma cells ob-
served on bone marrow biopsy, they were quite different at the
epigenetic level. Whereas REM2 as expected clustered with
MGUS samples, REM1 clustered with the relapsed samples,
suggesting the presence of epigenetic higher risk features that
were notappreciated withconventionalmethods.
FIGURE5. Decitabinetreatmentleadstogrowthinhibitioninmyeloma
Despite the continued progress and the improvement of treat-
celllines.Celllinesweretreatedwithdifferentdosesofdecitabinefor5d,
ment with newer drugs, drug resistance remains a big concern.
andproliferationwasassessedbytheMTSassay.Significantinhibitionof
growthwasseenaftertreatmentevenwithlowdosesofdecitabine(ttest, Strategiestousehighlyactivedrugsincombinationupfronthave
p,0.05).Shownisonerepresentativeofthreeexperiments. resultedinasignificantimprovementofprogression-freesurvival;
TheJournal ofImmunology 9
however, they have only had limited effect on overall survival. In contrast to early MM, relapsed MM shows predominant
Multiple studies in other cancers have shown association of hypermethylation.Althoughwecannotshowaclearreasonforthis,
hypermethylationphenotypeswithresistancetotreatmentviathe othershaveshownthatexposuretochemotherapycanleadtolarge-
inactivation of various cell cycles and other genes involved in scalegeneticandepigeneticchangesintumorsamples[reviewedin
chemosensitivity. The presented study identifies a relative hyper- (31–33)].Therelapsedcasesexaminedinourcohortwereheavily
methylationincasesofrelapsedmyelomascomparedwithnormal pretreated patients who have been exposed to multiple MM
plasmacells.Thisdifferenceisfurtherenhancedinthecomparison agents, including bortezomib. It is plausible that our relapsed
betweennewlydiagnosedMMandrelapsedMM(datanotshown) population is enriched with patients who have a higher overall
and consistent with the observation made by Walker et al. (18) degree of methylation, as hypermethylated phenotype has been
comparingplasmacellleukemia,averyadvancedhigh-riskstage associatedwithchemotherapyresistanceinmultiplecancers(34–
ofMM,withnewlydiagnosedMM.Atthispointitisnotpossible 36)andhasbeenlinkedwithresistancetobortezomib(37).Inthe
to say whether this hypermethylation phenotype is the result of current study, we have shown that low-dose decitabine signifi-
treatment or due to the biology of the disease. However, it sug- cantlyreducesviabilityofseveralMMcelllinesasmodelsoflate-
gestsinclusionofdemethylatingagentssuchasDNMTinhibitors stage disease. Low doses of decitabine have been shown to sig-
asaviabletreatmentoption.Invitrofindingsbyusandothers(27, nificantly reduce DNMT levels without the additional DNA
28) with representative myeloma cell lines using the DNMT in- damage activity (38,39). Ongoing longitudinal studies following
hibitor decitabine support this assumption, and clinical trials patients from diagnosis to relapse will hopefully answer the
testing this hypothesis are currently underway. questionastowhetherhypermethylationcanbedetectedearlyon
Inadditiontoglobalquantitativedifferencesinmethylation,we and whether it is associated with survival. If hypermethylation
also found differences in sites of aberrant methylation between weretobedetectedearlyandtopredictforshortprogression-free D
MGUS and myeloma. We observed that even though changes in survival,itwouldberationaltoaddademethylatingdrugsuchas o
w
MGUSinvolvedCpGislands,thelaterchangesseeninmyeloma decitabine totheinitial therapy. n
lo
preferentially occurred outside of CpG islands. Recent work has a
d
e
sbiemaiblaerrlryansthlyowmnetthhyaltacteydto/hsiynpeosmpertehsyelnatteodutisnidceanocfeCr,paGndiasslasnaydsstchaant Disclosures d fro
J.M.andS.B.S.areontheSpeakers’BureauxofCelgeneandMillenium. m
cover these loci are critical to discovering the full landscape of
Theotherauthorshavenofinancialconflictsofinterest. h
altered methylome of malignancies (16). Our novelfindings thus ttp
demonstratebothqualitativeandquantitativedifferencesbetween ://w
these subsets of disease. References w
w
genFeintiacllcyh,awnegeasl.soWperoovbisdeerviendsigthhatst DinNtoMthTe3Agenisessiisgnoiffitchaenstelyerpei-- 1. AMn.gDuhiaondoa,pkAa.r,,SJ..NAe.vTinusc,hBm.aBna,rClo.gAiec,hJ.arDy.aS,hKa.ugShanlteesr,syC,.JrG.,aasnpdarAe.ttPoo,tFti..Z20h0a9n., .jim
Geneexpressionprofilesoftumorbiologyprovideanovelapproachtoprognosis m
ducedinmyelomaandisaberrantlymethylatedearlyinbothMGUS andmayguidetheselectionoftherapeutictargetsinmultiplemyeloma.J.Clin. u
n
andmyeloma.DNMT3Aisanimportantmethyltransferasethathas Oncol.27:4197–4203. o
beenshowntobemutatedinalargeproportionofAMLcases(17). 2. ZS.hYana,ccFo.,bYy,.JH.Suaanwgy,eSr,.BC.oBlluar,inJ.gtPo.nS,teetwaal.rt2,0I0.6H.aTnhaemmuroal,ecSu.laGrucplatas,siJfi.cEatpisotneionf, l.org
WedidnotfindanyDNMT3Amutationsinourcasesandshowed multiplemyeloma.Blood108:2020–2028. b/
thatthisenzymecanbeaberrantlymethylatedincanceraswell.It 3. BJ.eSrghsaauggehln,esPs.y,LJ.r,.2W00.5.MC.ycKliunehDl,dyFs.reZghualant,ionJ.:aSnaewayrleyr,anBd.uBniafrylionggiep,atahnod- y g
isgenerallyacceptedthatDNMT3AandDNMT3Bareresponsible geniceventinmultiplemyeloma.Blood106:296–303. ue
s
for the establishment of methylation, whereas DNMT1 ensures 4. Zhan,F.,B.Barlogie,V.Arzoumanian,Y.Huang,D.R.Williams,K.Hollmig, t o
maintenance of methylation, thus guaranteeing the faithful trans- Mex.prPeisnseiodna-Rsiogmnaatnu,reGo.fTbreicnoigt,nFm. ovannoclRohneael,gMam.mZoapnagtahryi,eevtidael.nt20in07m.uGlteinpele- n M
missionofDNAmethylationmarksfromonecellgenerationtothe myelomaislinkedtogoodprognosis.Blood109:1692–1700. a
next. However, this longstanding view is being challenged, and 5. Figueroa, M. E., S. Lugthart, Y. Li, C. Erpelinck-Verschueren, X. Deng, rch
P.J.Christos,E.Schifano,J.Booth,W.vanPutten,L.Skrabanek,etal.2010. 3
newer models suggest a role for the de novo enzymes DNMT3A DNA methylation signatures identify biologically distinct subtypes in acute , 2
andDNMT3Binmethylationmaintenance(29).Infact,recentdata myeloidleukemia.CancerCell17:13–27. 0
1
show that DNMT3A mutations impart an adverse prognosis in 6. Alvarez,H.,J.Opalinska,L.Zhou,D.Sohal,M.J.Fazzari,Y.Yu,C.Montagna, 9
E. A. Montgomery, M. Canto, K. B. Dunbar, et al. 2011. Widespread hypo-
AML (30). Our data show that dysregulation of this methyl- methylationoccursearlyandsynergizeswithgeneamplificationduringesoph-
transferase can also be seen in myeloma and adds to the impor- agealcarcinogenesis.PLoSGenet.7:e1001356.
7. Oda,M.,J.L.Glass,R.F.Thompson,Y.Mo,E.N.Olivier,M.E.Figueroa,
tance of this gene in hematologic malignancies. Analysis of the
R. R. Selzer,T. A. Richmond,X. Zhang, L. Dannenberg,et al. 2009. High-
publiclyavailabledatafromWalkeretal.(18)confirmsasignifi- resolutiongenome-widecytosinemethylationprofilingwithsimultaneouscopy
cantly higher level of methylation for newly diagnosed MM numberanalysisandoptimizationforlimitedcellnumbers.NucleicAcidsRes.
37:3829–3839.
comparedwithnormalsamplesandaveryhighmethylationstatus
8. Chapman, M. A., M. S. Lawrence, J. J. Keats, K. Cibulskis, C. Sougnez,
of MGUS samples with a strong trend toward significance (data A.C.Schinzel,C.L.Harview,J.P.Brunet,G.J.Ahmann,M.Adli,etal.2011.Initial
notshown).However,generalizationofthisdataislimiteddueto genomesequencingandanalysisofmultiplemyeloma.Nature471:467–472.
9. Khulan, B., R. F. Thompson, K. Ye, M. J. Fazzari, M. Suzuki, E. Stasiek,
the low number of normal and MGUS samples included in that M.E.Figueroa,J.L.Glass,Q.Chen,C.Montagna,etal.2006.Comparative
study.Unfortunately,duetosamplelimitation,wewerenotableto isoschizomerprofilingofcytosinemethylation:theHELPassay.GenomeRes.
16:1046–1055.
obtainRNAforgeneexpressionprofilesfromthesamplesusedto
10. Zhou,L.,J.Opalinska,D.Sohal,Y.Yu,Y.Mo,T.Bhagat,O.Abdel-Wahab,
perform the methylation analysis. Although we are not able to M.Fazzari,M.Figueroa,C.Alencar,etal.2011.Aberrantepigeneticandgenetic
directly correlate methylation status and gene expression, the marksareseeninmyelodysplasticleukocytesandrevealDock4asacandidate
pathogenicgeneonchromosome7q.J.Biol.Chem.286:25211–25223.
findingofhypermethylationandunderexpression(comparedwith
11. Thompson,R.F.,M.Reimers,B.Khulan,M.Gissot,T.A.Richmond,Q.Chen,
normal plasma cells) seen in two different datasets, respectively, X.Zheng,K.Kim,andJ.M.Greally.2008.Ananalyticalpipelineforgenomic
as well as the DNMT3A expression effect on survival suggest representationsusedforcytosinemethylationstudies.Bioinformatics24:1161–
1167.
a role for DNMT3A in myelomagenesis that warrants further in- 12. Subramanian, A., P. Tamayo, V. K. Mootha, S. Mukherjee, B. L. Ebert,
vestigation.IfDNMT3Aexpressionandmethylationplayarolein M.A.Gillette,A.Paulovich,S.L.Pomeroy,T.R.Golub,E.S.Lander,and
J.P.Mesirov.2005.Genesetenrichmentanalysis:aknowledge-basedapproach
the progression of MGUS to MM, requiring treatment is the
forinterpretinggenome-wideexpressionprofiles.Proc.Natl. Acad.Sci. USA
subject ofan ongoinginvestigation. 102:15545–15550.