tumor and stem cell biology cancer research pml-rarα and...

Tum

PMAcu

DeepaScott

Abst

Intro

Acuthe bt(15;17cytic lgeneRARαresultcalledmiceloid-spplasialeukemgenetitencygestiveIt h

zymestargetand tr(7). Ad

AuthorReprodLaboraDepartmof Cal4DeparTexas M

Note:Resear

CorresFrancis415-47

doi: 10

©2010

Cance8792

Published OnlineFirst September 21, 2010; DOI: 10.1158/0008-5472.CAN-08-4481 Published OnlineFirst September 21, 2010; DOI: 10.1158/0008-5472.CAN-08-4481 Published OnlineFirst September 21, 2010; DOI: 10.1158/0008-5472.CAN-08-4481

Canceresearch

or and Stem Cell Biology

L-RARα and Dnmt3a1 Cooperate in vivo to Promote

R

te Promyelocytic Leukemia

Subramanyam1, Cassandra D. Belair1, Keegan Q. Barry-Holson3, Haijiang Lin4,
C. Kogan2, Emmanuelle Passegué3, and Robert Blelloch1
ractThe

ically iical doin vivomarrotransp

anscriptioditionally

s' Affiliatiouctive Scietory Medient of Me

ifornia attment of Oedical Bra

Supplemench Online (h

ponding Auco, 513 Pa6-2838; Fax

.1158/0008-

American A

r Res; 70(

DownloadDownloadDownload

PML-RARα oncogene is the central effector of acute promyelocytic leukemia (APL). PML-RARα phys-nteracts with epigenetic-modifying enzymes including DNA methyltransferases (Dnmt) to suppress crit-wnstream targets. Here, we show that increased expression of Dnmt3a1 cooperates with PML-RARαto promote early lethality secondary to myeloid expansion and dysfunction in primary mice. Bone

w cells from these mice cause leukemogenesis with a shortened latency and a higher penetrance onlantation into irradiated recipients. Furthermore, leukemic cells overexpressing PML-RARα anda1 display increased methylation at a target promoter compared with PML-RARα or Dnmt3a1 controls.
Dnmt3
Our findings show a cooperation between the PML-RARα oncogene and the Dnmt3a1 enzyme in vivo and thatDnmt levels can be rate limiting in APL progression. Cancer Res; 70(21); 8792–801. ©2010 AACR.

also indeace(10), alenceRARαAPL

superpterestDnmtationand Pfor Dnof leuabilityPML-Rexpresin enhkemogmarropients

duction

te promyelocytic leukemia (APL) is characterized byalanced reciprocal chromosomal rearrangement), which results in the juxtaposition of the promyelo-eukemia gene (PML) and the retinoic acid receptor α(RARA) to form the oncogenic fusion protein PML-(1). This is accompanied by a block in differentiationing in an overproduction of immature myeloid cells,promyelocytes, in the bone marrow (2, 3). Transgenicexpressing PML-RARα under the control of the mye-ecific promoter, cathepsin G, develop myeloid hyper-, with around 15% to 30% of mice progressing toia after a latency period of 9-12 months (4). Additionalc changes cooperate with PML-RARα to reduce the la-and increase the penetrance of APL development, sug-of a requirement for secondary cooperative events (5, 6).as been shown that PML-RARα recruits epigenetic en-such as DNA methyltransferases (Dnmt) to specificgene promoters, resulting in CpG island methylation

nal repression in cultured human APL cells, in vitro experiments reveal that PML-RARα

greateobtaincells fmethysults sto proopmen

Mate

MoushCG

miceDnmtthe re

ns: 1Institute for Regeneration Medicine, Center fornces and Department of Urology, 2Department ofcine, and 3Institute for Regeneration Medicine,dicine, Division of Hematology/Oncology, UniversitySan Francisco, San Francisco, California; andpthalmology and Visual Sciences, The University ofnch, Galveston, Texas

tary data for this article are available at Cancerttp://cancerres.aacrjournals.org/).

thor: Robert Blelloch, University of California at Sanrnassus Avenue, San Francisco, CA 94143. Phone:: 415-476-1635; E-mail: [email protected].

5472.CAN-08-4481

ssociation for Cancer Research.

21) November 1, 2010

Research. on February 1, 20cancerres.aacrjournals.org ed from



teracts with other epigenetic proteins such as histonetylases (8, 9), the methyl-CpG–binding protein MBD1nd the Polycomb repressive complex 2 (PRC2) to si-target genes, including those normally regulated by(11).cells can be forced to differentiate in the presence ofhysiologic doses of all-trans retinoic acid (ATRA). In-ingly, treatment of APL cells with a combination ofinhibitors and ATRA seems to enhance the differenti-of APL cells, suggesting a cooperative role for DnmtsML-RARα in APL maintenance (7). However, a rolemts or other epigenetic enzymes in the developmentkemia has not been shown. Therefore, we tested theof the epigenetic enzyme Dnmt3a1 to cooperate withARα in inducing APL in mice. We predicted that over-sion of Dnmt3a1 along with PML-RARα would resultanced silencing of PML-RARα targets and enhance leu-enesis. Indeed, transplantation of cells from the bonew of PML-RARα+Dnmt3a1 mice into irradiated reci-resulted in the development of leukemia with a

r penetrance and shorter latency compared with cellsed from PML-RARα mice. Additionally, leukemicrom PML-RARα+Dnmt3a1 mice displayed enhancedlation at a target gene promoter. Together, our re-how that PML-RARα and Dnmt3a1 cooperate in vivo
mote oncogene-specific target methylation and devel- t of APL.
rials and Methods

e strains-PML-RARα (4) mice were crossed to Rosa26rtTA

(12). Progeny of these crosses were bred to the TRE-
3a1 mice (13) to obtain triple transgenic mice andlevant controls. Mice were maintained on 2 mg/mL
19. © 2010 American Association for Cancer19. © 2010 American Association for Cancer19. © 2010 American Association for Cancer

http://cancerres.aacrjournals.org/



doxycdrinkiprime

ForRosRosRos

ForhCGhCG

ForColAColBColC

MicCalifoaccord

SurviMic

288 daof this

MethyTot

modifdiummenteevery

TransCon

sourcesplit d

FigureDnmt3acoopertransgehCG-PMTRE-DnProgenhCG-PMTRE-Don doxwater fshowinWT, DnPRhommice mcontainof age.a periodthe singdied. Pmousecompaand spl

PML-RARα and Dnmt3a1 in APL

www.a

Published OnlineFirst September 21, 2010; DOI: 10.1158/0008-5472.CAN-08-4481

ycline supplemented with 10 mg/mL sucrose in theirng water from 3 weeks of age onwards. The followingrs were used for genotyping:

Rosa26rtTA:a 26a: AAAGTCGCTTCTGAGTTGTTATa 26b: GCGAAGAGTTTGCCTCAACCa 26c: GAGGGGAGAAATGGATAT

hCG-PML-RARα:Fwd: GGCCTGACCTCATCCCATAGRev: GCCCTTTTCCCCATCCTAGG

TRE-Dnmt3a1:: GCACAGCATTGCGGACATGC: CCCTCCATGTGTGACCAAGG: GCAGAAGCGCGGCCGTCTGG

e were bred and maintained at the University of
rnia at San Francisco (UCSF), and their care was inance with UCSF guidelines.
water

een.

acrjournals.org

Research. on February 1, 20cancerres.aacrjournals.org Downloaded from

val analysise of the noted genotypes were followed over a period ofys, with percentage of survivors calculated at the endperiod.

lcellulose colony formational bone marrow cells (5,000) were cultured in Iscove'sied Dulbecco's medium–based methylcellulose me-(Methocult M3100; Stemcell Technologies) and supple-d as previously described (14). Cells were replated7 days on fresh methylcellulose medium.

plantation studiesgenic recipient mice were irradiated using a cesiumirradiator with lethal (1,200 rad) dose delivered in a

ose 3 hours apart and were given antibiotic-containing
for at least 4 weeks after irradiation. Mice were injected
6
immediately after irradiation. For i.v. injections, 2 × 10 donor
1. Overexpression of1 with hCG-PML-RARαates to shorten survival ofnic mice. A, generation ofL-RARα, Rosa26rtTA, andmt3a1 mice and controls.y of crosses betweenL-RARα and Rosa26rtTA;

nmt3a1 were maintainedycycline (Dox)–containingrom 3 wk of age. B, graphg survival percentage ofmt3a1, PR, PR+Dnmt3a1,, and PRhom+Dnmt3a1aintained on doxycycline-ing water starting at ∼3 wkSurvival wasmonitored overof 288 d. Asterisk indicatesle Dnmt3a1 mouse thatostmortem analysis of thisrevealed normal myeloidrtment in both bone marrow

Cancer Res; 70(21) November 1, 2010 8793

19. © 2010 American Association for Cancer


bonecongeand incells wof CD4transpcontaing frewere p

BisulfWe

versio3 μg oDNA wrestricsulfitecleanuufactuward:AACAmoter

the TOcomppande

Cell sFor

or sple(FBS).FcγR,CD3,(PE),(APC)at app20 miwashewasheHBSSsion.cell pfor an

Figurecompamarker(b) HSC(Gr1+/M(CD34+

differenanalysi

Subramanyam et al.

Cance8794


marrow cells were mixed with 300,000 Sca1-depletednic spleen cells, resuspended in a volume of 100 μL,jected into the retro-orbital plexus. Donor and recipientere distinguished by expression of different allelic forms5 (CD45.1 versus CD45.2). Throughout the experiment,lanted recipients were maintained on doxycycline-ining water. Round 1 transplants were performed us-sh bone marrow cells, whereas round 2 transplantserformed with cryopreserved cells.

ite sequencingisolated genomic DNA from spleen cells. Bisulfite con-n was performed as previously described (15). Briefly,f genomic DNA from spleen were digested with EcoRV.as purified using phenol-chloroform extraction aftertion digestion followed by denaturation with NaOH. Bi-conversion was carried out overnight at 55°C followed byp using the Promega Wizard Cleanup kit (using the man-rer's protocol). PCR was performed using primers (for-GGTTTGGTTAGGAATAGGAGAGTAGA; reverse:

ce in the percentage of cells in each compartment is calculated based ons was performed using Kruskal-Wallis test. *, P < 0.05.

ACCCTACAAAAACCTTCAAC) to amplify the RARβ pro-. PCR products were cloned into the PCR2.1 vector using

lyzedselect

r Res; 70(21) November 1, 2010


PO-TA kit (Invitrogen) and transformed into chemicallyetent bacteria. Individual colonies were picked, ex-d, DNA extracted, and sequenced using the same primers.

taining and flow cytometryflow cytometry, single-cell suspension of bone marrowen cells was prepared in HBSS + 2% fetal bovine serumThe following mouse antibodies, c-kit, Sca1, CD34,Flk2, Gr1, Mac1, Ter119, B220, CD19, CD8, CD4, andconjugated with fluorophores FITC, phycoerythrinPacific Blue, Cy7PE, Cy5PE, and allophycocyanin, were used for staining. These were added to cellsropriate concentrations and incubated on ice fornutes in the dark. Biotinylated antibodies wered and incubated with streptavidin-Cy7PE. Cells wered and resuspended in a final volume of 300 μL of+ 2% FBS with propidium iodide for dead cell exclu-Events (30,000) were collected for analyzing matureopulations, whereas 1,000,000 events were collectedalyzing stem cells using a BD LSRII. Data were ana-

llowing formula: fold difference = (Exp %)/(Control %). Statistical

2. Myeloid expansion in mice expressing PML-RARα and Dnmt3a1. Graphs represent fold differences in percentage of cells in mice analyzedred with their experimental control (either WT or Dnmt3a1) from five independent experiments. Based on the expression of specific cell surfaces, cells of the bone marrow and spleen were classified into the following populations: (a) hematopoietic stem and progenitors (KLS+; c-kit+/Lin−/Sca1+);s (c-kit+/Lin−/Sca1+/Flk2−/CD34−); (c) MPPs (c-kit+/Lin−/Sca1+/Flk2+/CD34+); (d) myeloid progenitors (MP; c-kit+/Lin−/Sca1−); (e) granulocytesac1+); (f) myeloid precursors (Mac1+/c-kit+); (g) B cells (B220+/CD19+); and (h) T cells (CD4+ or CD8+). MP cells were further subdivided into CMPs/FcγR−), granulocyte/macrophage progenitors (GMP; CD34+/FcγR+), and megakaryocyte/erythrocyte progenitors (MEP; CD34−/FcγR−). Fold

the fo

using FlowJo. Forward and side scatter were used togated cells for analysis.

Cancer Research



HistopSter

10% bufore em

OxidaInfla

collatevestedand re(300 μ(3 μL5 minuaddedwere iin FL1

Resu

BuildiDnm

moter

APL cDnmtducibltetracyTRE-Dversecontroand PMspecifitransgwe weRARαstrategmice (Dnmtwere dtheseDnmtone oAll mi

Tabana

Mou

21827958.9156226

237

218273254254254

272

254

254

NOTAbbcrysorga


www.a


athologynum, spleen, kidney, liver, lungs, and heart were fixed inffered formalin solution. Sternums were decalcified be-bedding in paraffin. Sections were stained with H&E.

tive burstmmation was induced by injecting 1 mL of a 3% thiogly-solution intraperitoneally. Peritoneal exudates were har-72 hours later by lavage with PBS. Cells were spun downsuspended in 1.5 mL HBSS + 2% FBS. Cell suspensionL) was aliquoted into each tube. Dihydrorhodamine-123of 29 nmol/L) was added and incubated at 37°C fortes. Phorbol myristate acetate (60 μL of 20 μg/mL) wasper tube and incubated at 37°C for 15 minutes. Samplesmmediately analyzed on a FACSCalibur, and the shiftfluorescence was determined using FlowJo.

lts

ng of an inducible Dnmt3a1–PML-RARα APLmodel
t3a forms a complex with PML-RARα at target pro-
s, resulting in their transcriptional silencing in humanbeginnwere i

le ary y of ns r froly

se Ge ge plee BM

.16 Dn 12 ONR −

.10 Dn 33 − −PR+ 40 ONR +++

.5 PR+Dnmt3a1 193 ONR m++

.1 PR+ 01 ONR m++

.4 PR+ 21 ONR m++

.4 PR+ 85 m+ m++

.6 P 32 m+ m++

.2 P 66 m++ m++

.13 PRhom 162 m+ m++

.1 Rhom 46 m+ m++

.2 Rhom 33 m+ m++

.6 Rhom 66 m+ m++

ns could not be retrieved.

acrjournals.org


ells (7). To determine whether PML-RARα and3a cooperate to induce leukemia in mice, we used an in-e system, whereby Dnmt3a1 is under the control of thecline-inducible promoter element (TRE-Dnmt3a1; ref. 13).nmt3a1 mice were crossed to mice expressing the re-tetracycline-controlled transactivator, rtTA, under thel of the ubiquitous promoter, Rosa26 (R26-rtTA; ref. 12),L-RARα (PR) under the control of the human myeloid-

c promoter, cathepsin G (hCG; ref. 4), to generate tripleenic mice PR+rtTA+Dnmt3a1 (PR+Dnmt3a1; Fig. 1A). Asre unable to determine the copy number of the PML-allele from these crosses, we later established a breedingy by which we obtained only homozygous PML-RARαPRhom mice). Mice that carried the R26-rtTA and TRE-3a1 alleles in addition to the homozygous PML-RARαesignated as PRhom+Dnmt3a1. Controls included instudies include mice carrying both R26-rtTA and TRE-3a1 alleles (Dnmt3a1 mice) and mice carrying eitherr none of the alleles [wild-type (WT) mice; Fig. 1A].ce were maintained on doxycycline-containing water
ing at 3 wk of age. Dnmt3a1 mRNA and protein levelsnduced, on the administration of doxycycline, only in
e inc in t uno ypic
1. Summsis
of histopatholog
orga etrieved
1

m mic

iver

−−m+−++

m+

−m+++

m+m+

m+

−

9. © 2010 A

luded

idne

−−

ONRONR−

−

−−−−−

−

−

Cancer

merican A

he imm

Lungs

−−−c+m+

ymphc++m+c++−−m+−

m++ymphc+

ymphc+m+

ymph

Res; 70(21) No

ssociation fo

phenot

eart

−ONR−−−

−

−ONR−−−

ONR

−

vember 1, 201

r Cancer

no.
notype A S n L K y H
mt3a1 1
mt3a1 1 Dnmt3a1 1 m
Dnmt3a1 1
m l +
Dnmt3a1 1

Dnmt3a1 1
Rhom 1 Rhom 1 m
P
+Dnmt3a1 1 l +
P
+Dnmt3a1 1 l +
P
+Dnmt3a1 1 l +
c+.11 PRhom+Dnmt3a1 162 m+ m++ m+ − m++ −

c++

E: −, normal.reviations: BM, bonemarrow; m, myeloid expansion/infiltration [+, low; ++, intermediate; +++, extensive (leukemic)]; c, eosinophilictals (+, low; ++, intermediate; +++, extensive); lymph, lymphocyte infiltration (+, low; ++, intermediate; +++, extensive); ONR,

0 8795


mice t(Suppnot re(Suppl

PML-RlethalMic

the hCperiodDnmt3wemomorbiDnmt(n = 21on doearlyPRhomthan tmice (where251 daPML-R

MiceDnmtthe bTo

expresand pand spPRhowell-cmatop(HSC)proge(GMPture gCD4 Tseen i(Fig. 2showedecrePRhomFig. S2

ureice

an srrowm

lysiunoH&magesmyaretratiow;ensilyzed is shown in Tables 1 and 2.

Subramanyam et al.

Cance8796


hat carried both Rosa26rtTA and TRE-Dnmt3a1 alleleslementary Fig. S1A and C). Induction of Dnmt3a1 didsult in an alteration in PML-RARα expression levelsementary Fig. S1B).

ARα and Dnmt3a1 cooperate to enhanceity in micee expressing only PML-RARα under the control ofG promoter develop fatal leukemia with a long latencyand a low penetrance (4). To determine whethera1 cooperates with PML-RARα to enhance fatal leukemia,nitored control and experimentalmice for changes in theirdity and mortality over a period of 288 days. WT (n = 6),3a1 (n = 10), R (n = 13), PR+Dnmt3a1 (n = 28), PRhom), and PRhom+Dnmt3a1 (n = 24) mice were maintainedxycycline-containing water. Lethality was observed asas 51 days for PR+Dnmt3a1 mice and 43 days for+Dnmt3a1 mice (Fig. 1B). This was significantly earlier

he lethality seen for PR mice (264 days) and for PRhom211 days). No early lethality was observed in WT mice,as a single mouse died in the Dnmt3a1 group at
ys. These data show that Dnmt3a1 cooperates withARα to shorten survival of mice.
otherbroad



expressing PML-RARα alone and PML-RARα with3a1 display similar levels of myeloid expansion inone marrow and spleendetermine the cause of morbidity and mortality in micesing PML-RARα and Dnmt3a1, immunophenotypingopulation analysis was performed on bone marrowleen cells isolated from control Dnmt3a1, PRhom, andm+Dnmt3a1 mice. Cells were separated based onharacterized cell surface markers defining different he-oietic populations including hematopoietic stem cells, multipotent progenitors (MPP), common myeloidnitors (CMP), granulocyte monocyte progenitors), megakaryocyte erythrocyte progenitors (MEP), ma-ranulocytes, immature myeloid cells (c-kit+/Mac1+),cells, CD8 T cells, and B cells. Dramatic changes weren several of these populations across the mutant mice; Supplementary Fig. S2). Specifically, the bone marrowd a significant increase in the fraction of GMP andase in the fraction of MEP and B cells in PRhom,+Dnmt3a1, and PR+Dnmt3a1 mice (Supplementary, top; Fig. 2). No significant change was observed in the

Figof morgmafromanaimmwith×40imawith(m)infil+, lextana

bone marrow cell populations.er spectrum of changes with inc

19. © 2010 American Associa

3. Histologic characterization. Representative images ofections of heart, kidney, bone, liver, spleen, and lungsice included in the survivals or those included in thephenotypic analysis, stainedE. All images are taken atgnification. Representativeof normal tissue and tissueeloid expansion/infiltrationshown. The degree ofon is represented as follows:++, intermediate; +++,ve. Summary of all mice

The spleens showed areases in the HSC, MPP,

Cancer Research

tion for Cancer


immacompplemeof organd spmice,Fig. 3+Dnmsimila(Suppilar mor wit

Mice ebeforTo

were dmia, hwere r

Of thefrom tspleentrationkidneyly descmicespleenbonemice,mice tmild-tand speratecells.myeloliver, h

Tab

Mou

485291273

291291

291302

291

291272291

272291

291

291

234234

NOTAbbcrysorga


www.a


ture myeloid, and mature myeloid populations, ac-anied by a decrease in the percentage of B cells (Sup-ntary Fig. S2, bottom; Fig. 2). Histologic examinationans revealed a myeloid expansion in the bone marrowleen of PRhom, PRhom+Dnmt3a1, and PR+Dnmt3a1consistent with the flow cytometric analysis (Table 1;) . Bone marrow cells from PRhom and PRhomt3a1 mice cultured in vitro on methylcellulose showedr proliferation and differentiation into myeloid cellslementary Fig. S3). Together, these results show a sim-yeloid hyperplasia phenotype in PML-RARαmice withhout Dnmt3a1 overexpression.

xpressing PML-RARα and Dnmt3a1 die at a stagee the development of leukemiadetermine whether PR mice overexpressing Dnmt3a1ying earlier due to premature development of leuke-
istopathologic analysis was performed on organs thatetrieved from mice shortly after death (Table 2; Fig. 3).
velopmgle PR

le ary gy eve

se Ge e a leen

.5 P +++

.16 P +++

.12 P +++++

.10 hom ++

.8 hom ++

.14 PRhom+ Dnmt3a1 118 m+++

.3 hom +

.18 hom ++

.2 hom NR

.6 hom +

.12 hom NR

.5 PRhom+ Dnmt3a1 251 m++

.15 hom NR

.9 hom NR

.5 hom +

ns could not be retrieved.

acrjournals.org


five PRhom mice that died, organs could be retrievedhree of these mice. Two of them had grossly enlargeds with complete loss of splenic architecture and infil-of immature myeloid cells into the lungs, liver, and. Together, these findings are consistent with previous-ribed features of APL observed in the hCG-PML-RARα(4). However, the third mouse had a normal-sized, although a myeloid expansion was observed in themarrow and spleen. In the PRhom+Dnmt3a1 group ofhistologic examination was performed on 10 of the 22hat died. Nine of them had normal-sized spleens with ao-moderate myeloid hyperplasia in the bone marrowleen. The livers of these mice showed a mild to mod-perivascular infiltration of well-differentiated myeloidOne mouse had an enlarged spleen with immatureid infiltration, as well as myeloid infiltration in theeart, lungs, and kidney, consistent with leukemia de-
ent. We were able to retrieve organs from only a sin-+Dnmt3a1 mouse, which displayed mild-to-moderate
m lud the al a
2. Summ of histopatholo of organs retri d from
M

+++++++++++

m+++++

++

R++++

m++++

R

++

19. ©

ice inc

er

++++++

+

m++++

++

R

+

m+R

R

+

2010 Ame

ed in

ne

−+−

−−

m+++−

−

NR−−

−−

NR

−

Cancer Re

rican Ass

surviv

ung

++++c+

+++mphc++m++m+mph+++mph++++++c+++++++c+m+mphc++phc++mph

s; 70(21) No

ociation for

nalysis

art

++

++

vember 1, 201

Cancer

no.
notype Ag t death (d) Sp B Liv Kid y L s He
Rhom
229 m m+ m m − m+ m+ m Rhom
Rhom
278 m284 m m+ m+
m
−m+
c
c c − PR
PR
+ Dnmt3a1+ Dnmt3a1
202 m287 m

mm+

−m

cly
+ −
+ m+−
PR + Dnmt3a1 180 m m m
ly
+
PR
+ Dnmt3a1 265 m m+ m cly + − c
PR
+ Dnmt3a1 275 O ON ON O c − − PR
PR
+ Dnmt3a1+ Dnmt3a1
263 m265 O

mm+

−m
m −
c
−− PR + Dnmt3a1 222 O m+ ON
ly
+
PR
+ Dnmt3a1 277 O ON ON O lym ++ −
PR
+ Dnmt3a1 188 m m m ly + − c++
.9 PR + Dnmt3a1 251 m+ m++ m+ − lymph+ −

.10 Dnmt3a1 251 − − − − − −

E: −, normal.reviations: BM, bonemarrow; m, myeloid expansion/infiltration [+, low; ++, intermediate; +++, extensive (leukemic)]; c, eosinophilictals (+, low; ++, intermediate; +++, extensive); lymph, lymphocyte infiltration (+, low; ++, intermediate; +++, extensive); ONR,

0 8797


myeloperivaThe scompThereseemethan lInde

PRhomin theand mtheirconsisconditand mlow fre(Tabletweenthe exation idid ingranulaffect(Supp

accentnia prconsoresultithe pnanimaopmen

TransPML-RdevelTo

of DnmlethalvestedPRhomthe cenors wspleniitored200 dahypot

ureDn

diatpaiatece

homkeminstee dotyicatectedbesecbe

ipients used in round 2.

Subramanyam et al.

Cance8798


id hyperplasia in the bone marrow and spleen and ascular infiltration in the liver of mature myeloid cells.ingle Dnmt3a1 mouse that died was found to haveletely normal histology of all organs examined.fore, Dnmt3a1 overexpression in the PR backgroundd to be inducing early lethality due to causes othereukemia.ed, histologic examination of organs from preleukemic+Dnmt3a1 mice also revealed inflammatory infiltrateslungs, composed of mature granulocytes, lymphocytes,acrophage-like cells filled with eosinophilic crystals incytoplasm (Supplementary Fig. S4; Tables 1 and 2),tent with crystalline macrophage pneumonia. Thision has been observed in C57Bl/6 mice, 129Sv mice,ice deficient for p47phox (16–18). It also occurs atquency and with less severity in PML-RARα alone mice1). The exact etiology of the observed cooperativity be-PML-RARα and Dnmt3a1 is not clear. However, givenpression of PML-RARα in the myeloid lineage, an alter-n myeloid cell function is likely in part responsible. Wefact observe a hyperresponsiveness in PML-RARαocytes, althoughDnmt3a1 could not be shown to further
this in an assay for release of reactive oxygenmetaboliteslementary Fig. S5). Overall, it seems that Dnmt3a1
kemogto leu



uates a propensity to crystalline macrophage pneumo-eexisting in PML-RARαmice. Importantly, the infiltrateslidated large parts of the lungs in the PR+Dnmt3a1 mice,ng in a dramatic reduction of airway space. Therefore,eumonia is the likely cause of early lethality in thesels, possibility not permitting enough time for the devel-t of leukemia.

plantation of bone marrow cells expressingARα and Dnmt3a1 results in the

opment of leukemiaovercome the difficulty of assessing the contributiont3a1 to leukemogenesis due to the early nonleukemic

ity, we performed bone marrow transplants. We har-bone marrows of littermate mice of PRhom and+Dnmt3a1 genotypes at 6 to 7 months of age and used

lls to reconstitute lethally irradiated recipient mice. Do-ere at a similar stage of preleukemia as determined byc composition (Supplementary Fig. S6). Mice were mon-for leukemia development and survival over a period ofys (Fig. 4; Supplementary Fig. S7). Consistent with thehesis that Dnmt3a1 cooperates with PML-RARα in leu-

FigwithirracomirradwithPRLeuagaThrgenindinjenumthenumrec

enesis, two PRhom+Dnmt3a1 dkemias with features of APL in

19. © 2010 American Associa

4. PML-RARα cooperatesmt3a1 to cause leukemia ined recipients. Kaplan-Meierrative survival analysis ofd congenicmice transplantedlls from either PRhom or+Dnmt3a1 donors is shown.ia-free survival is plotteddays after transplantation.onors were used for eachpe. Numbers in parenthesisthe number of recipientsper donor, with the first

r representing round 1 andond number round 2. Singlers indicate the number of

onors rapidly gave riseall recipients, whereas

Cancer Research

tion for Cancer


recipiwith lscoredthat reture. Tinto oRepet

the trPRhomdid nocontriWe te


www.a


ents of PRhom donor marrows developed leukemiasonger latencies (Fig. 4). Leukemia development wasby the presence of infiltrating immature myeloid cells

sulted in the enlargement and loss of splenic architec-he splenic phenotype was accompanied by infiltration
ther organs such as the lungs, liver, kidneys, and heart. of bot
acrjournals.org


ansplantability and rapid lethality of cells from+Dnmt3a1 mice, whereas cryopreserved PRhom cellst cause leukemia in recipients even at 200 days despitebution to long-term hematopoiesis (data not shown).sted a third cryopreserved mouse at 6.5 months of age
h PRhom and PRhom+Dnmt3a1 genotypes and observed
ition of this experiment using frozen cells replicated leukemia only in the PRhom+Dnmt3a1 recipients. Collectively,

Figure 5. Methylation status of the RARβ promoterin spleen cells of donor and leukemic recipients. A, schematicof the region of the RARβ promoter. Each line with a circleat the end of it represents a CpG dinucleotide. Numbersindicate the positions relative to the transcription startsite (+1). B, bisulfite genomic sequencing of the RARβpromoter for representative transplant donors andleukemic recipients. Each circle represents a CpGdinucleotide. Open circles are unmethylated and blackfilled circles are methylated CpGs. Each row representsan individual sequencing run. The percentageofmethylatedCpGs is denoted next to each sample.

19. ©


2010 American Association for Cancer


mice ra 91%arisingtransplater ti(Fig. 4)followmanyFig. S7with PTheseindeed

PML-RdownThe

wouldmethygenesiPML-Rthe DNRARβcreasethe prleukemylationaloneTheseeratestargetin mecells hcounte

Discu

Ourcoopesis. InDnmt3infiltraprogreouslyinfreqdramatranspcells pshortePR+Dn

PML-Rfore, wase catargetThe

PML-RActivatrancethe PMin APLapproin conthe roseveraation,gestinleukeDnmtmor fHowepathwDnmttein. Tedge tepigenspecuof sevtancegenes

Discl

No p

Ackn

We tthe oxidmice, TBlelloch

Grant

Univgrant K

Theof pageaccorda

Refe1. de

ThtionRA

2. Be

Subramanyam et al.

Cance8800


eceiving PRhom+Dnmt3a1 cells developed leukemia withpenetrance (21 of 23 transplanted mice), with leukemiasfrom all 3 donors. In contrast, only 18.5% (5 of 27) ofmicelanted with PRhom cells died of leukemia and at muchme points, with only 2 of 3 donors giving rise to leukemia. Indeed, several PR-alone transplants survived the 200-day-up (Supplementary Fig. S7), and of those that died,died of nonleukemic causes (compare Supplementarywith Fig. 4). In contrast, only one mouse transplantedRhom+Dnmt3a1 cells died from nonleukemic causes.findings show that Dnmt3a1 overexpression doescooperatewith PML-RARα in promoting leukemogenesis.

ARα and Dnmt3a1 cooperate to methylate astream target in leukemia developmentincreased leukemia in PR+Dnmt3a1 recipient micesuggest that these two proteins cooperate in vivo tolate downstream targets and hence promote tumori-s. To test whether Dnmt3a1 promoted methylation ofARα targets, we used bisulfite sequencing to analyzeA methylation status of the PML-RARα target genein spleen cells (Fig. 5A). The RARβ promoter showed in-d CpG methylation in leukemic recipient mice relative toeleukemic donors. More importantly, PRhom+Dnmt3a1ic mice showed consistently higher levels of DNA meth-than their PRhom counterparts (Fig. 5B). Dnmt3a1

mice showed no evidence of increased methylation.findings show that overexpression of Dnmt3a1 coop-with PML-RARα to promote hypermethylation of itsgenes during the progression of leukemia. The increasethylation with time is consistent with the methylatedaving a competitive advantage over their unmethylatedrparts.

ssion

results show that inducible expression of Dnmt3a1rates with PML-RARα to promote APL leukemogene-primary mice, overexpression of PML-RARα anda1 resulted in early death secondary to an eosinophilicte that consolidated the lungs, leading to death beforession to leukemia. Such an infiltrate has been previ-described in the PML-RARα model (19) and was seenuently in PR-alone mice in this study. However, it wastically enhanced in the PR+Dnmt3a1mice. In secondarylant experiments, recipientmice receiving PR+Dnmt3a1rogressed to leukemia with higher incidence and
r latency than their PR-alone counterparts. Leukemicmt3a1 cells showed increased DNA methylation at a
ReceOnlineF

in acute promyelocytic leukemia encodes a functionally alteredR. Cell 1991;66:675–84.nnett JM, Catovsky D, Daniel MT, et al. Proposals for the classi-

ficCo

3. Begra

4. Gr



ARα target relative to leukemic PR-alone cells. There-e conclude that the levels of a de novo methyltransfer-n be rate limiting in the methylation of PML-RARαs and the progression to leukemia.requirement for secondary mutations in addition toARα for the development to APL is well documented.ting mutations in Flt3 or in K-Ras increase the pene-and shorten the latency of leukemia development inL-RARα mouse model (5, 6) and are commonly foundpatients (20–23). Accumulation of epigenetic marks at

priate targets could also function as cooperating eventsjunction with PML-RARα. Recent studies examiningle of epigenetic marks in human APL suggest thatl target genes are heavily decorated by DNA methyl-histone methylation, and histone acetylation, sug-g that significant epigenetic changes occur duringmia development (24). Previous work showed that3b could enhance the development of intestinal tu-ormation in an APCmin mutant background (13, 25).ver, this cooperation likely occurs through parallelays, as there is no evidence that the mostly nuclear3b enzyme interacts with the cytoplasmic APC pro-herefore, our work is the first example to our knowl-o test cooperativity in vivo between an oncogene andetic-modifying enzyme that physically interact. Welate that a deeper exploration into the developmenteral cancers will highlight the ever-increasing impor-of epigenetic enzymes in cooperating with onco-and tumor suppressor genes in cancer progression.

osure of Potential Conflicts of Interest

otential conflicts of interest were disclosed.

owledgments

hank S. Espejel-Carbajal and S. Sevcikova for technical assistance withative burst assay, Rudolf Jaenisch for sharing the Dnmt3a1-inducibleim Ley for sharing the hCG-PML-RARα mice, and members of thelaboratory for critically reading the manuscript.

Support

ersity of California Cancer Research Coordinating Committee and NIH08 NS48118 (R. Blelloch).costs of publication of this article were defrayed in part by the paymentcharges. This article must therefore be hereby marked advertisement innce with 18 U.S.C. Section 1734 solely to indicate this fact.

ived 11/24/2008; revised 08/02/2010; accepted 09/15/2010; publishedirst 09/21/2010.

rencesThe H, Lavau C, Marchio A, Chomienne C, Degos L, Dejean A.e PML-RAR α fusion mRNA generated by the t(15;17) transloca-

ation of the acute leukaemias. French-American-British (FAB)-operative Group. Br J Haematol 1976;33:451–8.
nnett JM, Catovsky D, Daniel MT, et al. A variant form of hyper-nular promyelocytic leukemia (M3). Ann Intern Med 1980;92:261.isolano JL, Wesselschmidt RL, Pelicci PG, Ley TJ. Altered myeloid
Cancer Research



dePMBlo

5. Chwitdis

6. Sois rSU

7. Dimeic t

8. Griacimia

9. LintheNa

10. VillMBA 2

11. Villco513

12. Bemeint

13. Lininsile

14. Akmo200

15. He

Mof

16. Homo20

17. Liudymo

18. Mumi

19. Maa da msio

20. CaFLleuLe

21. KiyasskemLe

22. Loin a

23. YoFLlodA s11

24. Mathe


www.a


velopment and acute leukemia in transgenic mice expressingL-RAR α under control of cathepsin G regulatory sequences.od 1997;89:376–87.an IT, Kutok JL, Williams IR, et al. Oncogenic K-ras cooperatesh PML-RAR α to induce an acute promyelocytic leukemia-likeease. Blood 2006;108:1708–15.hal J, Phan VT, Chan PV, et al. A model of APL with FLT3 mutationesponsive to retinoic acid and a receptor tyrosine kinase inhibitor,11657. Blood 2003;101:3188–97.Croce L, Raker VA, Corsaro M, et al. Methyltransferase recruit-nt and DNA hypermethylation of target promoters by an oncogen-ranscription factor. Science 2002;295:1079–82.gnani F, De Matteis S, Nervi C, et al. Fusion proteins of the retinoicd receptor-α recruit histone deacetylase in promyelocytic leukae-. Nature 1998;391:815–8.RJ, Nagy L, Inoue S, Shao W, Miller WH, Jr., Evans RM. Role ofhistone deacetylase complex in acute promyelocytic leukaemia.ture 1998;391:811–4.a R, Morey L, Raker VA, et al. The methyl-CpG binding proteinD1 is required for PML-RARα function. Proc Natl Acad Sci U S006;103:1400–5.a R, Pasini D, Gutierrez A, et al. Role of the polycomb repressivemplex 2 in acute promyelocytic leukemia. Cancer Cell 2007;11:–25.ard C, Hochedlinger K, Plath K, Wutz A, Jaenisch R. Efficientthod to generate single-copy transgenic mice by site-specificegration in embryonic stem cells. Genesis 2006;44:23–8.hart HG, Lin H, Yamada Y, et al. Dnmt3b promotes tumorigenesisvivo by gene-specific de novo methylation and transcriptionalncing. Genes Dev 2007;21:3110–22.ashi K, Traver D, Miyamoto T, Weissman IL. A clonogenic com-
n myeloid progenitor that gives rise to all myeloid lineages. Nature0;404:193–7.rman JG, Graff JR, Myohanen S, Nelkin BD, Baylin SB.
Ce25. Lin

sia

acrjournals.org


ethylation-specific PCR: a novel PCR assay for methylation statusCpG islands. Proc Natl Acad Sci U S A 1996;93:9821–6.enerhoff MJ, Starost MF, Ward JM. Eosinophilic crystalline pneu-nia as a major cause of death in 129S4/SvJae mice. Vet Pathol06;43:682–8.Q, Cheng LI, Yi L, et al. p47phox deficiency induces macrophagesfunction resulting in progressive crystalline macrophage pneu-nia. Am J Pathol 2009;174:153–63.rray AB, Luz A. Acidophilic macrophage pneumonia in laboratoryce. Vet Pathol 1990;27:274–81.rchesi F, Monestiroli SV, Capillo M, et al. Eosinophilic crystals asistinctive morphologic feature of a hyaline droplet nephropathy inouse model of acute myelogenous leukaemia. J Vet Med A Phy-l Pathol Clin Med 2003;50:103–7.llens C, Chevret S, Cayuela JM, et al. Prognostic implication ofT3 and Ras gene mutations in patients with acute promyelocytickemia (APL): a retrospective study from the European APL Group.ukemia 2005;19:1153–60.oi H, Naoe T, Yokota S, et al. Internal tandem duplication of FLT3ociated with leukocytosis in acute promyelocytic leukemia. Leu-ia Study Group of the Ministry of Health and Welfare (Kohseisho).

ukemia 1997;11:1447–52.ngo L, TreccaD, Biondi A, et al. Frequency of RAS and p53mutationscute promyelocytic leukemias. Leuk Lymphoma 1993;11:405–10.kota S, Kiyoi H, Nakao M, et al. Internal tandem duplication of theT3 gene is preferentially seen in acute myeloid leukemia and mye-ysplastic syndrome among various hematological malignancies.tudy on a large series of patients and cell lines. Leukemia 1997;:1605–9.rtens JH, Brinkman AB, Simmer F, et al. PML-RARα/RXR altersepigenetic landscape in acute promyelocytic leukemia. Cancer

ll 2010;17:173–85.
H, Yamada Y, Nguyen S, et al. Suppression of intestinal neopla-by deletion of Dnmt3b. Mol Cell Biol 2006;26:2976–83.



Correction

Correction: PML-RARa andDnmt3a1Cooperatein vivo to Promote Acute PromyelocyticLeukemia

In this article (Cancer Res 2010;70:21:8792–801), which was published in the Novem-ber 1, 2010 issue of Cancer Research (1), the grant support is incomplete. Thecorrected grant support is provided below:

Grant Support

University of California Cancer Research Coordinating Committee and NIH grant K08 NS48118 (R. Blelloch)and NIH grant R01 CA095274 (S.C. Kogan).

Reference1. Subramanyam D, Belair CD, Barry-Holson KQ, LinH, Kogan SC, Passegue E, et al. PML-RARa

and Dnmt3a1 cooperate in vivo to promote acute promyelocytic leukemia. Cancer Res2010;70:21:8792–801.

Published OnlineFirst March 29, 2011.�2011 American Association for Cancer Research.doi: 10.1158/0008-5472.CAN-11-0285

CancerResearch

www.aacrjournals.org 2805

2010;70:8792-8801. Published OnlineFirst September 21, 2010.Cancer Res Deepa Subramanyam, Cassandra D. Belair, Keegan Q. Barry-Holson, et al. Promyelocytic Leukemia

to Promote Acutein vivo and Dnmt3a1 Cooperate αPML-RAR

Updated version

10.1158/0008-5472.CAN-08-4481doi:

Access the most recent version of this article at:

Material

Supplementary

http://cancerres.aacrjournals.org/content/suppl/2010/09/21/0008-5472.CAN-08-4481.DC1

Access the most recent supplemental material at:

Cited articles

http://cancerres.aacrjournals.org/content/70/21/8792.full#ref-list-1

This article cites 25 articles, 8 of which you can access for free at:

Citing articles

http://cancerres.aacrjournals.org/content/70/21/8792.full#related-urls

This article has been cited by 4 HighWire-hosted articles. Access the articles at:

E-mail alerts related to this article or journal.Sign up to receive free email-alerts

Subscriptions

Reprints and

[email protected] at

To order reprints of this article or to subscribe to the journal, contact the AACR Publications

Permissions

Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)

.http://cancerres.aacrjournals.org/content/70/21/8792To request permission to re-use all or part of this article, use this link

Research. on February 1, 2019. © 2010 American Association for Cancercancerres.aacrjournals.org Downloaded from


http://cancerres.aacrjournals.org/lookup/doi/10.1158/0008-5472.CAN-08-4481

http://cancerres.aacrjournals.org/content/suppl/2010/09/21/0008-5472.CAN-08-4481.DC1

http://cancerres.aacrjournals.org/content/70/21/8792.full#ref-list-1

http://cancerres.aacrjournals.org/content/70/21/8792.full#related-urls

http://cancerres.aacrjournals.org/cgi/alerts

mailto:[email protected]

http://cancerres.aacrjournals.org/content/70/21/8792


tumor and stem cell biology cancer research pml-rarα and...

Documents