Role of Histone Deacetylase HDAC7 in B Lymphocyte Biology

Role of Histone Deacetylase HDAC7 in B Lymphocyte Biology

Lidia Román González

ADVERTIMENT. La consulta d’aquesta tesi queda condicionada a l’acceptació de les següents condicions d'ús: La difusió d’aquesta tesi per mitjà del servei TDX (www.tdx.cat) i a través del Dipòsit Digital de la UB (diposit.ub.edu) ha estat autoritzada pels titulars dels drets de propietat intel·lectual únicament per a usos privats emmarcats en activitats d’investigació i docència. No s’autoritza la seva reproducció amb finalitats de lucre ni la seva difusió i posada a disposició des d’un lloc aliè al servei TDX ni al Dipòsit Digital de la UB. No s’autoritza la presentació del seu contingut en una finestra o marc aliè a TDX o al Dipòsit Digital de la UB (framing). Aquesta reserva de drets afecta tant al resum de presentació de la tesi com als seus continguts. En la utilització o cita de parts de la tesi és obligat indicar el nom de la persona autora.

ADVERTENCIA. La consulta de esta tesis queda condicionada a la aceptación de las siguientes condiciones de uso: La difusión de esta tesis por medio del servicio TDR (www.tdx.cat) y a través del Repositorio Digital de la UB (diposit.ub.edu) ha sido autorizada por los titulares de los derechos de propiedad intelectual únicamente para usos privados enmarcados en actividades de investigación y docencia. No se autoriza su reproducción con finalidades de lucro ni su difusión y puesta a disposición desde un sitio ajeno al servicio TDR o al Repositorio Digital de la UB. No se autoriza la presentación de su contenido en una ventana o marco ajeno a TDR o al Repositorio Digital de la UB (framing). Esta reserva de derechos afecta tanto al resumen de presentación de la tesis como a sus contenidos. En la utilización o cita de partes de la tesis es obligado indicar el nombre de la persona autora.

WARNING. On having consulted this thesis you’re accepting the following use conditions: Spreading this thesis by the TDX (www.tdx.cat) service and by the UB Digital Repository (diposit.ub.edu) has been authorized by the titular of the intellectual property rights only for private uses placed in investigation and teaching activities. Reproduction with lucrative aims is not authorized nor its spreading and availability from a site foreign to the TDX service or to the UB Digital Repository. Introducing its content in a window or frame foreign to the TDX service or to the UB Digital Repository is not authorized (framing). Those rights affect to the presentation summary of the thesis as well as to its contents. In the using or citation of parts of the thesis it’s obliged to indicate the name of the author.

UNIVERSITATDEBARCELONA FACULTATDEFARMÀCIA PROGRAMADEDOCTORATENBIOMEDICINA2013

RoleofHistoneDeacetylase HDAC7inBLymphocyteBiology

MemoriapresentadaperLidiaRománGonzálezparaoptaraltítulodedoctorporla UniversitatdeBarcelona

EstetrabajohasidorealizadoenelGrupodeDiferenciacióncelulardelProgramade EpigenéticayBiologíadeCáncer(PEBC)delInstitutdeInvestigacióBiomèdicade Bellvitge(IDIBELL),bajolasupervisióndeDr.MaríaIsabelParraBola

Director Tutor Doctorando

MaríaIsabelParraBola EstebanBallestarTarín LidiaRománGonzález

Acknowledgements



Contents

Contents

Contents

Acknowledgements 5

ThesisAbstract 17

ResumendeTesis 21

PARTI INTRODUCTIONANDGLOBALHYPOTHESIS 25 @ R -

/^/56_ 8 9 / R /56 6/56 6/56 : -/567

Contents -=>9 -G< G 6/56 6/56 /

PARTIIOBJECTIVEANDMETHODOLOGY67 HG= = -/56R 6R 5/566 R 8/56R66 G _ 5/56 R 8/56_R R 5/56R6 >G /56=>6 =R

Contents

-/56 X/56@@ 5 G/56@@ 5 /56 @@ /56 X6/56R R -/56 R =/56 R

PARTIIIRESULTS 99 -/56R 5/566 RR 8/56R66 G_ 5/56 R 8/56

Contents

_RR 5/56R6 >G/56=>6 =R -/56 /566 X/56@@ 5 G/56@@ 5 /56 @@ /56@@ X6/56RR =/56 R 5/56R Rexvivo

PARTIVDISCUSSION,CONCLUSIONANDREFERENCES 147 6 -

Contents ANNEX1LISTOFFIGURESANDABBREVIATIONS207 Y 5GG

ANNEX2PUBLICATIONS223 YG 5 5

















ThesisAbstract

ThesisAbstract

5 ><G 556 5 _ G/5G _

: := 8 ^/56_ @ 5 G 5 /56G5 /56 95 6 R 5 5 R R 5 =>9 G5

G=5 67invitroGRG 5 in vivo G @@R

G/56 5 @ G GR^5 _  /565 /56

ThesisAbstract

R 6 /56 6 R R5R6 /56 _ 5 G@ @ 5 ^ 5_5G 5 G G 6G6=R@=5/56 =>96 5

/565 @@R /56 ^R_H/56R G@R5/56 =5/56 =>9G R5G/56 R

5 /56

ResumendeTesis

> X S5S5< S X 6S / 5 S X G S 3 < G G 8 G5 S S=_

>G= S Y ^/56_ S X S <_5 < > 5 /56 63=< S>5/56GR 9_53S R

: G= 5 < G= 67 6S 6 in vitro S 5 6S in vivo SS/56

: GS /56 S X R S5 < S _ 3 X X

_SSR ^5_/_5/56 5/56S6S3=> _ R6S /56 _S 3S5 S 5 S5 G 6 35 /56 =>96 S5 S

:/56X5 S /56 3 X ^X R_ < _/56G_ XR5_/56 S 35/56S=>96< = X R GX _/56S 3S

>S5/56

PARTI INTRODUCTIONANDGLOBALHYPOTHESIS

Introduction

Introduction

1. Blymphocytedevelopment

1.1.IntroductionandOverview

G < / ^/86_ R G /86 R ^Y=::_ ^6Y:_6Y: 5 @ ^<`_ ^`5 6 G5 6 =56/5_Y=:: ^6=:_5 G L ^X=:_ @L ^=>:_ X=: 55=>: @ ^=5 6 85 _ ^9_

5 5 G< GRag1Rag2Y==:5 6Y:^5_6Y:6 5665G_ R6- ^R _ ^ 6 =_5 5RR5:R G G6G

Introduction

6@R ^-@56H5_

Figure1i:Modelforhematopoiesis.Y RG^=:5_

1.2 .TranscriptionfactornetworksinthegenerationofBlymphocytes

5 R 65 @ > G 6 @ G 5 5

Introduction

G6@R ^56=56 Y5_

5Y=::`5-H85:=>96 ^9 _ IKAROS GIkzf1^X56/5_ R R G 6 ^6-6_ <5 5 8L8 <9G^8L8<9_=RL<-^=R L<-_ ^_ G G G65Igll1 ^_5G>9<5G^H<56 `5 6 85 _ `5-H8R G@ Y=:: ^X5 6 5 _ =5 G`5-H8RY=::G 55^56 <5_PU.1 Y=::GSfpi16 5 :G5 ^856`5_G :Y=::5_5 55<`5 G ^85 6 =`5 6 `5 _ MEF2CG^=_ <G=58RG6<5 _ ^X566=5_Y5G

Introduction

=>96 6 ^85 _ 5 =>96 @ 5 Ets1, Tcf7, Gata3, Pbx1, Il7r, Rag1, Ciita, Cd23, Cr1/Cr2 and Tnfsf45 =>96RG =>96R Y==: GR6Y:5 _6Y:^8R856G5_

Y 5 >55>95:56Y5^9_E2A_ RRGTcfe2a ^5 _ >5 _ 6 >95 :5RR6 GRag1,Rag2,Vpreb2,Vpreb3Cd19^56 56H-56856=B5_G >55G@RR5

G/R/^56565 _ EBF G >9 R 65Cd79a^Mb1_5Cd79b5Blk5VpreB1,VpreB2,VpreB3,CD19 5^Igll1)>959HHR5 6Y: ^H-5 6 5@G5 6 5@G5 6 =B5 _ >9R 6Y: /G R ^H-5 6 5 _ -5 G >9 6 5Ebf1

R X5 Ebf1

X G@ ^5G5 6 XV5 _ PAX5 : ;

Introduction

66 ^6G5 6 55 6 Y5 6 ==5 _ :5 56R 55:56G R 5 5 G G :5 cd195 Y>9R

RR6Y^56:568@565 _ 5 G G :5 5 G <5 G@Cd19,Cd21, Cd22, Cd23, Cd40, 6 =/6 ^/5 _ :5 G@ R5 URG^G<@566G565_ 5>9:5BCL11A. 6Y5@R< ^Y5 6 85 _ 6Y5 _56Y:5G 6 >95 :5 Y- ^Y5 _ -5 G G 6Y5Rag1Rag26Y5 ^_G56Y5@ G ^_ G^Y5_

R _ IRF4 IRF85 6 R6- 6 Rag1 Rag2

Introduction

^56=5_^9_IRF4IRF8GR < G <5 G G G RG^=5_-96 G -9 6 R ^5_RG^68-_55Y=:R 5 ^95 6 `5 6 85 _ -9R G5 Y=:R < ^=*@5 6 Y5 _ IRF8 G R ^>G56:G568565_-9 6 R^Y56=<5_ 5 -9 6 6Y 5 ^Y5 _ -9R RY^/@5 6Y5_

< R6- FOXO1^9_R5R6- 59HHG G 6 ^-5 6 85 6 =5 _ @ ^:`_ Y- 5 9HH5 6 5 9HH Y-6Rag1,Rag2, Sell ^ YR ^6Y__ Aicda ^5 6 5 _ 9HHR RR

Introduction

5_59HHR6= ^5_

9556YY=:R R ^X6_ 5 ^9_BCL6_ ^X6_G 5 R X6 6Y R G /56 R5 <6H- 5 6H-5 ^9=56-56:5_6YR X6 G R G ^9@5 6 5 _ 95 RR R (BLIMP1) ^_Y=:R6 G 6 : 5G666- =/6 <5 Y=: G@6@^SpiBId35 _5G^ 55 `5 `5 <5R:`65 85_5 G G ^85 _ =Y=:RG R^8568R85_

Introduction

Figure2i:TranscriptionfactorexpressionduringBlymphopoiesis. ^=-<5 _

1.3.TranscriptionfactormediatedgenesilencinginBlymphocytes

G G G

65 5 55 6 PAX55 6 5 G 5:56 :5 R ^/5 6 5 6 8G5 _ :5 G

Introduction

G ^6G5 _ X 6 R:5R:5R 5 :5 ^ Cd33, Notch1, Itgal Ly6d_5 ^ Ccl3 Ccl9_5 ^ Mpa21, Tes Vav3_5 @ ^ Myo11 _5 G ^ CmahAcadm_5G^Capn5:_ ^ Lmo25 Mef2c 8_ ^5 6 8G5 6 :5_5G @:5R_ 5:5G R5:5 @ 5 /56 :5 ^-RR5_

5 EBF1 5 >9 5 R Cebpa ^ 6L>:_5 Pu.15 Sfpi15 Id2 = 55@55 :5^:G5656G5 _ G >95 6GU 6=R689-XR689-^:G5_ @5G>9G G<RPu.1 Cebpa.

Introduction

IRF8 G @ -9 6 > G ^:5 >Y_5 -9 G ^-95 -9 -9_5 ^>5 <95 =_ ^5 6 `5 6 -G5 6 <5 6 5 _ = G -9R GG X 6 R -9R G -9 5 R -9R 5 R 5 565655 R < =5 G -95 G -9 /86^565_

= 5 MEF2C _ 6 R ^8R85 6G5_8R8etal.=>96 Y=::5G G =>96@@R58R8etal. =>96R Y=:: G 5 Y=:: XU 5 Y=:: 9 5 =>96 R @ R 6L>: 5

Introduction

5Y=::5=>96 <^8R85_

5 5 G RR

1.4.ReprogrammingofBcells

G G 6R 8 et al. >6 Drosophila melanogaster 6 Antennapedia^Antp_5R@5 G 6 ^85 _ 5 5 etal.R R6G^=H_ GG ^5_Y5X G G 6 6L>: 6L>: 6L>: :55 R6G:5 Mac1GR5 ^5 6 Y5 _ 8G_5 G / et al. 5R5GG G 6 6L>: @@R :5 ^/5 _ G 5 X G

Introduction

RR@ Q R G 6 RG 6L>: ^5_=5G @ G R@5GR66L>:6L>: ^-5 _  GG5GG< 65<5G R ^-3:R 5 C/ebp 6 5 ^`5 _ -5 G G 6L>: 6 R = R = 5 @ 5L5 5 ^85_

@ 5 G <

Introduction

2. HistoneDeacetylases

2.1. Introductionandoverview

@5<5<G 5 ^/5/5/5/5< 5 /R/ /5R/_ G <5 ^Y5_:R^:=_ 6 G G <5 ^-5_

:= 5 5 5 5:RG5 G_5 5 G5 5 R G R < R G 5 ^=5 _ -5 -<5 ^-<5_ 6 G 5 5 ^ XG5 6 5 _ 65 -<5 ^-<5_ 6G5 5-<5G5<^:@5_

5 R G<7^/5_ ^/56_9/5/56 5G/5

Introduction

/566^56 5_< 56GG : 5 ^9_^/G5 _

Figure3i:Proteinacetylation.: G^/5_ ^/56_^555_

R5/56 R 5 R <R @ 8 R ^585855_ ^ RG R_5 ^:565_

/56 @

Introduction

/56 G 5 5 G5 5G^5_

2.2. Structure

< R @ ^=5 _ 65 /56 R ^R5_

= /56 5 G ^=5_6/56^/5655 _5-^_^95_6 6/565G5^/56555_ G^/56_5G/^ _^=5_665@^8-55555 _ 8 ^ _ ^5 6 5 _6 6  ^/56_5 GG/56<^9_^X5_

ClassIHDACs/5655G 6/56< 6R

Introduction

R_^X?5_-5G/56 ^5_

ClassIIHDACsG<G6 /565 /565 5 5 6 R ^9 _ 6 @5 5 5G555G 5 /56 5 5 956:/:

Introduction

^X5695656 85_

ClassIIIHDACs,sirtuins,G^8-R _ <8-55 8- < 8- 5 ^R :X5 6 85 _ ^9_ 8 <5U < 5 R 6 5 R 5^-=5_8 <5@5<G <559HHG5`5 5-G5^-<_5 G5<5G ^=56565656-=5_

95 /56 class IV HDACs G ^9_ 6 G5 5 @ @ /565<5 5 G @R5 H 5 5-/5G5^5_

Introduction

Figure4i:Thehistonedeacetylasefamily. GG ^=R5_

2.3. Function

5 /56 GG<

Introduction

=>95 5 ^=`5 6 55 6 65 6 65 6 5 _ 8-9 -< G ^56656=56565_ G/56G 5/56/56 ^95 _ 6 G 8/ /:5 <5 @^5_R5/56 R5 ^58585_^RG R_^:5_

2.3.1. Histonedeacetylation

/ G < <5 := <5L L G ^`<568G<5_/ G ^/5_ ^/56_5 5 5 ^85_^9_5/56/` ^R5 _ /56 /` G <5 ^/@5 _ /56R @ /` /`5 5 <5 GR

Introduction

G@ ^@5 _ 6 /565 /565 5 5 G /56 ^-=5 _ /56 /56 G 5 /56 = < G R ^=5_558-8-558-5 6 5 8- /`5 5 G 8- 8- /` @ ^<5 6 5 _6 8- /`5 GG6^X5_95/565 /565/`L` R65 ^Y5_

5 G5 6 5 5 G ^55_

Introduction

Figure5i: Role ofhistoneacetylationandchaperonesin nucleosome assembly. ///G//6 659R5R</R/ <5 /5R/ H G 5 G ^/56_^=8G<5_

2.3.2. Deacetylationofnonhistoneproteins

X R /566/56 GR5 R ^ 85_ ^ RG :><_ ^X5 6 Y5 6 :56:5_

H 6 R 5GG8-8 G5 8-

Introduction

^Y56<5_^9_8-5 65 /565 85 85 YR 85 5 <55^/5_G 8- < G5 85 G <5 G 85 ^5 _ :>< R /565 5:><G L6:R ^:659_ 6 5 G8-/56:><^:5_5 6R RG G /56 5 /56 G X8`G^/5665_H6/56 R8-8-5 ^X/_ 5 6 G6 ^65_ ^85 6 YG5 6 /5 _ /565 G   ^_5 ^85 6 5 6 5 _ /56 X55RG ^5_=58-G <9R+5 <9R+^`5_

Introduction

Figure 6i: SIRT1 p53 pathway regulators. 8 5 G 8- ^=Y5_

2.4. HDACsinphysiologyandpathology

/ /56 6 <5 G 5 G 6 65 /56 G G =5R/56 5 656 G/56/56

Introduction

5 @5 G5 5 ^`5 _6/56R56 ^ 6 @5 G5 @ 5 _ ^ 5 _ ^65_

5/56 6/565 6 @ ^5 _ > 6 /56 G ^:@5 _5 6 /56 ^H5 _ 5/565 G55G 65/56G^/_R5 /56^`5_/ GR5 5X55<9+^5_=/5 /565^<5_5 6/5^85_5/ ^95_5/56 G _ /56 G/5G/566 /56 G@G 5 /56 @G

Introduction

3. ClassIIaHDACs

3.1. Introductionandoverview

5 /565 /56 ^/565 5 _ _ 95 6 R 6 R6 @5 5 5 G5 5 5G585/56 955 R R G

3.1.1. ClassIIaHDACsinphysiologyandpathology

/56 5@@RG /56= @ /56 @ G G 5/566G -6 ^5 _ Y5 6 =>9GG/56 =>96_G^55_/56 65 -R^56 6565665_=5G /565R ^`5 _ G / ^/_ / ^=@5_

Introduction

= /56 L /56 G R 5 /56 ^65 _ 5 /56 @ 5G@GG G=>9^Y5_/56 G 65 6 -< 8=55 -< 5 G ^/5 6 `5 _ 5 G G -R /565 R /56 ^Y5 _ = 5 /56 G @ 5 @ ^5_5G G/5659Y6>R@G ^9Y:_ R G^:_5 ^5_

/5666U6U55 ^`5 6 _5 6 `5 _ G5 /56 /56G5 = G /565 =>9 5 ==: ^ 6 G_5/56 ^65_^9_/56 G5 -< ^ G _ ^565_

Introduction

Figure7i:HDAC7andMEF2controlMMP10expression.= ==: G /56 =>9 ^>6_ 5 /56 G =>95 ==: G /565 ==: 6 5 6 6 5 G^=65_

95 /56 G@^56:56 85 _ 5 /56 =>9 5 ^/G5 _ = /56 =>9 ^5 _ 5 /56

Introduction

6 /56R G^==5685_5 G/56 9H:5 /56R @ @ ^565_

6 /56 6 5 @ G5 5G55 R^:56=5_ /56GG /565 5 LR @ ^6=`_ : @ ^:`_5 R@^5=:`_ ^:56=X56XX5_5<5RG =>9 6=` /56 /56 6=` 5 /56 =>9R/56 5=>9 =>9 /56 =>9 5 5 6 /56 ^_G6=`5=>96 56/565=>9 6 G 6=` ^Y5 _/56GR G =>9 X5 @^=X5_5G

Introduction

/56R ^6=5_

3.2. RegulationofclassIIaHDACsfunction:Nterminaldomain

95 /56^Y56_56 /5_/56 <G<^=`565 _ 5 R 5G_55 ^=5_^9_

Figure 8i: Class II HDACs domain organization. 6 < 956:

Introduction

/:5G=5 GR < 5 ^=65_

3.2.1. RegulationoffunctionviaMEF2interaction

6 5 /56 G 8/56G<55 _R <5 G ^=5 _ 5 G < =>9 /56_ G^Y56_56/5_^9_=>9 G 6 =58R6 =>9 <5RG 5 <5 <5 G ^85 6 @5 6 =`5 _ =>9 6 =>9 6 5 5 5 5 G G R =>9 /56 G /56 =>9 6=>9^55665665 6 Y5 6 <5 6 5 _ G /56 =>9 G 5 =>9 ^@5 6 =`5 _ /56G6=`/56=>9 =>9 ^=`5 6 Y5 _ 5 G

Introduction

R^:/:_5 /565=>9 ^5 6 `<@5 _ 5 6 G =>9 /565 G@ G5/56 =>9G ^65_

3.2.2. Regulationofsubcellularlocalizationbyphosphorylation

95^8-9_-65 <5 ^65 _ /5 5 /56 G

Introduction

Figure9i:Regulationofsubcellularlocalizationbyphosphorylation.>6 @ :`5 6=` =5:` /566 5::5::5R <G5R< ^=65_

3.3. ClassIIaHDACscatalyticactivity

R /56 /^RQ_G GR/56^Q__R ^=5_/5G/565 <^956/5_ 56@/56GG 5 <55G^Y5

Introduction

_ 5 G /565GGG ^YG5 _ 9 65 /56 -= 5GGR6^5 _5/566R 5

3.4. ClassIIaHDACsinlymphocyteBiology

/56 /56 /56/56G@ G6^_ R 9H: 8@5 /56 /56 6 9H: G ^5 6Y5_-6/569H:6 ^5_/56 /5666U6UGR< 566 5 5 /56 Nur77, ^`5 6 _5 6 :5 _ 6L6GR56-

Introduction

G /56 G /565 6-5 ^`5_=5L@:`/56 R^6-_ 6/565RNur77, ^:56_5_5::L ::5RRG/56 ^=5 6:5_85R5:`:`_ R 6 /56 /56 ^6-_ ^=5 _ -5 G 6 = 6:R_ /56>55>996RG ^Y5_95G /56/566G@ ^5YY_5 6 ^=5 _ =5 G : 5 /56 G ^-5 _ H56G G /56 GR G /566G

Hypothesis

Hypothesis

5 G 5 G GR 5 5 R 6 5 @6 8@ 5 < G R 65 R G/5G 5R GR

55Ggenetranscriptionalrepressionis critical to acquire a particular cell fate in the hematopoietic system. G 6 G 6

PARTII

OBJECTIVESANDMETHODOLOGY

Objectives

Objectives

= to investigate the role of the histone deacetylase HDAC7 in B lymphocyte Biology 5 _ G 6 7 95 in vitro GR6^ _G^5_55in vivo6^@@R_

Specificobjectives:

1. -/56R7 5 /56 6 R 8 /56 R6 6 G_ 5/56R 8/56_R R 5 /56 R6 > G /56 =>6=R

2. -/567 /56 6 X /56 @@ 5 G/56@@

Objectives

5 /56@@ X6/56R =/56R

2.7.5/56RR exvivo.

Methodology

Methdology

1. RoleofHDAC7duringthereprogrammingofpreBcellsinto macrophages

/56 R R@ G 6 6^5_

1.1.Cellcultureandestradioltreatment

R /59Y 6 -:= QG5=5LL 6^/59Y=86RX9:R6L>:_6 ^/59Y =86R6R6L>: _ 6 -:= Q G 5 = 5 L L -5 A6 =>= Q G 5 = 5 L L R G GG^5_95 =R=YR =689

1.2. AnalysisofClassIIaHDACsexpressionduringthereprogrammingofpreBcells intomacrophages

/56 R < 6G-R_:6-5GRG

Methdology

RT–qPCRexperiments

-<56RR5 /59Y-556G<6^‹_ <5 < / 6 <5 - ` ^5 5 _ -R_:6- 8- X = ^-_ :6- < Y6 8^-_

Primers

_G7

Rag1FW X56565X6566XX66XXX Rag1RW 55X6566X666565XX66 Pax5FW X66XXX5XX556XX5 Pax5RW XXX6X65XXX6X555X5 C/EbpFW 5665X6666566XX5X C/EbpRW X5XX6XX5X5XX55X6X Csf1rFW X66X66665XX56555 Csf1rRW 65X65X5X65665X65X5X Hdac4FW X6656XX5X66X5 Hdac4RW 5XX65XXX66X5 Hdac5FW 5X665XXX56X66 Hdac5RW 5X6656X5X5XX566X Hdac7FW XXX65XXXX6565X Hdac7RW 5X6XX565XX5XX6655X Hdac9FW 66665XXX5X5XX5X Hdac9RW 655X656555665X65X

Westernblotanalysis

6R R/59Y -56 :YG^QR5=>55=:8_

Methdology

G ^65 - = _ 88R:5X> G6G

Antibodies

5R/56^/R_5R/56^6R_5R/56^/R_5R>5^R_5 R:5 ^6R_ R-< ^_ 8 6< 5R`5-H8 ^G_ 5G6 R=>96 ^6_:686RRG^_58R 5

1.3. Study of the effect of HDAC7 reexpression in C10 cells and its possible interference with the acquisition of the macrophage gene transcription programduringcellularreprogramming

/56_ RR 56G R

Plasmids

=86RR/56G/56R9<5 :6-<5R/56=86R^6_

Retroviralsupernatantgenerationandcellulartransduction

9 =86R =86RR/56 @ :R> RR96R=86 6R/56 5 6

Methdology

L

Westernblotanalysis

6R=866R/56RR 6:YGG 6GG

Antibodies

5R/56^/R_86<6R9 ^=_RRG^_58R5

Microarrayexperiments

6R=866R/56R ^ _ -<5 6 G < :6- -<5 G<56^=X:=_ - X 9 56 6>Y ^<_ GX>HGGX8>

Microarraydataanalysis

56 6>Y G@ 5 < 5 @ :; ^ _ :6; ^X5 6`5_856 ^_ G5 < _ G 6 7 G@ :;6 < :;6 :;6:;‹6^

Methdology

_ G @ :‹=; ^ 6-_^`5_ 6 ^>_ G 6 5 @Y< ^85 _ :G G 5 G 6 ^6<_ 65 G 59-R_5Y^Y96_ RY96 G6/56R>6 = G X>H G ^<X8>_^8R85_< @ 8 G@ G <5<_ ^L_ R5 Y96`

Functionalandpathwayenrichmentanalysis

9 6 G X H^XH_^7LL_6>>=Y`>XX G55 ^_G^:_5^_5^66_^=9_ `>XXG@XHL X ^_-R= =/G9-^9-_

Methdology

RT–qPCRexperiments

6R=866R/56RR -<56G<6^‹_<5< /6<5-`^55_-R _:6-GG

Primers

_G7

Cxcl10FW 666X5X66556666XX55 Cxcl10RW X65X6XX566X66X6 Fcgr1FW 5X6656XX5XX565X65XX Fcgr1RW X56XXXX56655X656XX5X Ifi35RW 56X6X6XX5XX6X5XX5 Ifi35FW 5X5XX565XX65XXX6565 Ifi47FW 66XX56X5X6666XX Ifi47RW 6X55X6566665XX55XX6 Tlr3FW 66X6XX6X55XX55X5 Tlr3RW XX555656666XXXX5X5566 Tlr9FW X66X6665XX65X5XX Tlr9FW 666X56XX5X55665XXXX Il18FW XX66X56656X565566X6 Il18RW XX6XXXX656XX656 69 56XX5XX66656X5X665 6- XX6XX556X5XX5X56X6 :69 X66XXX5XX556XX5 :6- XXX6X65XXX6X555X5

1.4. AnalysisofthephysiologicalfunctionofHDAC7inpreBcells

< /56 R 5 RR 6 5 G -<5^-<5_@@/56G/59Y

Methdology siRNAdepletion

-<5 R /56@@6/59Y -<5 Y -<5=5 -^_

RTqPCR

-<56-<5G-R_:6-

Primers

_7

Hdac7FW XXX65XXXX6565X Hdac7RW 5X6XX565XX5XX6655X ItgamFW XX666X666X5XXX65X5 ItgamRW 5X5X665656X66566XX6 Fcgr1FW 5X6656XX5XX565X65XX Fcgr1RW X56XXXX56655X656XX5X Ccl3FW 565X66XX55X56656X66 Ccl3RW 65XX5555X56566XX6XXX

1.5. StudyoftheinteractionofHDAC7withsequencespecifictranscriptionfactors inpreBcells

/56 _R R 5 R 6 <65 R ^6:_ 6 /56 =>9 G

Methdology

CoimmunoprecipitationandWesternblotanalysis

6 :Y G ^Q R5 = >55 = :8_ G ^65 - = _ G /565 88R:5X>  G 6 G G

Chomatinimmunoprecipitationsassays(ChIP)

6 R 6 G Q - _=5 G^QR5 = >55 = >X55 = /5 / _5 G 6 ^ = <65 = >55 = >X55 = /5 / _5 6: G G^Q885QR5=<65=>55=>X55= /5/_G5 -^665Y_R ^_R^_6= :5j/56=>96GG 85RG@ X G 6 : G^Q885Q65QR5 = <65 = >55 = >X55 = / / _5 G ^Q 885 Q65QR5=<65=>55=>X55= //_5G^=Y65Q65Q<:R5=>55 =>X55=/5/_5G^=>55=>X55 = /5 / _ 6 G

^Q 885 = </6H_ -5 @ 6

= <6 <5 7 65 7655

Methdology

/HG@ 6<5^j_G= _:6- qPCR

_:6-6GG

Antibodies

5R/56^/R_86<R =>96^6_:68

Primers

_ 6: 6 7

Cxcl10

:6-5

97X66565X5XX5XX665X -755XX556X55X55XXX55X555XX 97X5666X666666 -766X6XXX6X5XXX 975665X56665X65XX5X -765X665565XX5665XX 975565X5X5X65XX6565X -756555X55X56556555X6566 9765555X56XX65XX5566 -7X66555XX6655X655X 976655X665X65X6556 -7XX65XXX5XX55X5X6

Itgam

:6-5

9756XXX655XX666

Methdology

-75566566X5X5655XX6 9756X6X5X56XX566 -7XX6665666X66 97X6XXXX5XXX5XX65XX5X -7XX65XX6665XX6XXX 976565XX65X56X66X5XX -7X6X566566XX665566 976XX5XX55XX6565X5XX -7X6566565X5656X5XX

Fcgr1

:6-5

976X5X55XX5X55XX55X6 -75565XX565X5XX55566 97XX56566X656X6 -76X5XX665XX66X 97X5566XX6XX556X -75XX555X6XX6566 975XX555X6XX6566 -7X5566XX6XX556X 975666X566X55X656 -76XX5X5XXXXX55XX

VennDiagrams

G R /56 R G=>96Y=:: 8R856R_

Methdology

1.6. AnalysisoftheeffectsofHDAC7reexpressiononthefunctionalpropertiesof thereprogrammedmacrophages

/56 R6 5 @ 6  < 6 @ 6 = 5 G

Flowcytometry

6R=86 6R/56 R 5= G =R 6 ^ :_ =R 6 6 X 9 6 ^@ 6_ < G 9^85_

Phagocytosisassays

6R=866R/56G= / E.coli ` ^_

LPSinducedinflammatorycytokines

6R=86 6R/56 R GY:8^ jLY 8_ -<5 65 <5 -R_:6- GG:__

Quantificationofproinflamatorygenes

Methdology

_R< 7

TnfalphaZ-Z9 6X666X656X556 TnfalphaZ-Z- X5X5X5XXX5XX665 Il1bZ-Z9 665555566XXX66XX Il1bZ-Z- X6XX66X6XX5X Il6Z-Z9 X5XX5566566665565X566 Il6Z-Z- 55XX65656XX65565

1.7. ElucidationofthecontributionoftheinteractionofHDAC7withME2Cand itscatalyticactivityontherepressionofMac1

G /56 R 5 6/565=>9G

Plasmids

=86RX9:R/565 =86RX9:R/56^@=>9_5 =86RX9:R /56^`5L`5_5 =86RX9:R/56^/5_5 =86RX9:R/56^T_ =86RX9:R/56^T_ G /56R9 <5:6-6<5R/56R =86RX9:^6_

Retroviralsupernatantgenerationandcellulartransduction

9 =86RX9:5 =86RX9:R/565 =86RX9:R /56^@=>9_5 =86RX9:R/56^`5L`5_5 =86RX9:R/56^/5_5 =86R X9:R/56^T_=86RX9:R/56^T_

Methdology

@:R>TR 6=86RX9:5=86RX9:R/565=86R X9:R/56^@=>9_5 =86RX9:R/56^`5L`5_5 =86RX9:R/56^/5_5 =86RX9:R/56^T_ =86RX9:R/56^T_ R<6=RG5 GG

2. RoleofHDAC7inBlymphocytedevelopment

/56 5 in vivo, 5 /56@@/56

Miceandanimalcare

/566:L@ ^65 _ @ G > < H ^68585_5=GR6@LU^/G@5_ G = - ^=6 :@ >5 9G5 X_ 5 5 5 G - ^>YY_ 5 8

DNAextractionfrommicepunch

<55Y:G ^=<H/6=>5_A6H5

GenotypingPCR

Methdology

<565:6-7

Mb1Creki/+:

=6-^_7Mb1Cre 96 6 5   A6  5 <:^=_ 5= A6  =6^=_ = A6  6 5 =8H Q A6  :9^=_ 5= A6  :-^=_ 5= 5 _<5 5 /

8-^_7

G G `G G

HDAC7LOXP/KO:

=6-^_7HDAC7/loxP 96 6 5   A6  5 <:^=_ 5= A6  5 =6^=_ 5= A6  6 5 =8H Q A6  :85 5 ^=_ 5= A6  :Y 5 ^=_ 5= :85 ^=_ 5= 5 _<5 5 /

8-^_7

G G 96 G

Methdology

<G G

2.1. GenerationofHDAC7conditionalknockoutmiceinBcellprogenitors

8 /56R G ^65_@@@ 6 G ><H^8_5G6R5

Experimentaldesign

HDAC76L@5mb1R6@LU /56 6 5 G 6 R HDAC7loxP/ko mb1R6@LU HDAC7+/loxP;mb1Creki/+ HDAC7+/ko;mb1Creki/+ 5 G HDAC7loxp/ko;mb1Creki/+ @@ HDAC7+/ko;mb1 Creki/+

2.2. Analysisofthehematopoieticcellpopulationprofileinthebonemarrowof theHDAC7knockoutmice

HDAC7loxp/ko;mb1Creki/+ G G G G<@ 6 HDAC7+/ko;mb1Creki/+ 6 <GX^X5@R6_ <9

Bonemarrowextraction

Methdology

 G 5G55 YG7:8UQU=>5 ^:865985=>5_  GGG^ G_G  6 G @ jG  9GG  5GG G@^< _

 -[56`G^

j>5=9Y/H=/_8 -[YGG  6@ Cellstaining

 6YG^L_  59@^jL_G@  8  5j:5^Q855Q85<:8_UG  @  8  :8  8  5j:5Q:95:8

<7

Methdology

:R^U6U=R_ :R^U6R=R_ ^U6R=UR_ =^U6R=UU_ X^XU6GU_ =^XR6GU_ >^U_

2.3. Analysis of the hematopoietic cell population profile in peripheral lymphoid organsofHDAC7knockoutmice

/56_ G 5 < 5 5 G HDAC7loxp/ko;mb1Creki/+ HDAC7+/ko;mb1Creki/+ 6 <G<9

Spleenandthymusextraction

 : R :8 @  :  -  :8  9G:8  8  -[56`G  8  -[YGG

Methdology

6@

Peripheralbloodextraction

: G /566L@6mb1R6@LU /56UL@6mb1R 6@LU65 G G G >5 H 6556`GG

Cellstaining

6 G G G <7

^U6R=UR_ =<^=_^U6U6Y6R_ 9^9H_^U6Y6U6R_ ^_^U6Y6U6U_ G^:_^6U6U_ 86^8:_^6U6R_ 86^8:_^6R6U_ X^XU6GU_ =^XR6GU_

2.4. ImmunohistochemistryofspleensfromHDAC7mutantmice.

8HDAC7loxp/ko;mb1Creki/+HDAC7+/ko;mb1Creki/+ 6QA65G6 - 9 5 =^/_8G@Q RG G G A6 G

Methdology

R9LR68G G8Y688: ^Y_

2.5. GeneexpressionprofilingofHDAC7deficientproBcells

 6 6G/56R HDAC7loxp/ko;mb1Creki/UHDAC7+/ko;mb1Creki/+G

RNAextraction

-<5 6 R < ^_ G<569X6 9 ^9X6_ - ^-_ G

MicroarrayvalidationbyRTqPCR

GG -R_:6- G _ 7

Klf4F 5X5565X6656665656X Klf4R XXX55XX66X66X JundFW 66X5X5XXXX6655XX JundRW 555XXX6XX556XX5656 CebpzFW 6X66655565XXX565X5 CebpzRW 556X6XX6666X Malt1FW X6655XX665X55666 Malt1RW XX5X5566X6XX6555 SykbFW 6666X55X655556556X6 SykbRW 5XX5XX5X55X566X65 CD69FW 56556655X5XXXX66 CD69RW 6X665X5556X65X66

Methdology

Map3k8FW X55XX6XX65556X566 Map3k8RW 66X66XX6665555X5 NfkbidFW 5X566X55XX665666 NfkbidRW 65656565X6XXX5X65 Usp7FW 55XXXX5XX6555XXX Usp7RW XX65XX65XX6X55 JunbFW 65XX6X6566X55X65 JunbRW X6XXXXX56X5655 NfkbizFW X56XX5X5XXX5656X NfkbizRW 66566X555XX6566X CrebzfFW 6666X65X5XX6X CrebzfRW 656655XX665X6656 FosbFW 666X5X55X5X56565666 FosbRW 66X66X55X6X56X Egr1FW X5X6566X566565X5X6 Egr1RW X55X6XX665X55XXX5 TET2FW X6X666X5XX5X TET2RW XXX655X6X6X55X5 C/EbpFW 5665X6666566XX5X C/EbpRW X5XX6XX5X5XX55X6X CD28FW XX5XX66XXX66566X5 CD28RW 65XXXX6XX56X6X6555 Mdm2FW 55X65X655X566XX656 Mdm2RW 666XXXX6X6X

VennDiagrams

GR/56R R R /56 R 6R_

Methdology

2.6. RecruitementofHDAC7tolineageinappropriategenesinproBcells

/56 R6:6R  6 G G G5 R = G ^6-_ 6  6: 6G

PurificationbyantiPEmultisortkit

 6  8  -7jYLjYG  =6GA6  -7jYLjYR:>GL  =6GA6  RL  8  -7jYL

Magneticseparation(LScolumns)

 -G  5  6G6G  -G

Removalofmultisortmicrobeadsusingmultisortreleasereagent

 6  5jYL

Methdology

 =6GA6  RGL  8  -7jYL  5jYL  5YG  =6GA6

MagneticlabelingwithCD43microbeads

 6  8  -7jYLjY6GL  =6GA6  RGL  8  -7jYL

MagneticSeparation(LScolumns)

 -G  5  6G6G6^6-U6R _  -G^6-U6U_

2.7. MolecularmechanismsunderlyingHDAC7repressioninproBlymphocytes

95 6 < /56R exvivo. 5 6GHDAC7loxp/ko;mb1Creki/+

Methdology

HDAC7+/ko;mb1Creki/+RG= GGR8 @ ^YR5 8695 95 6R YR_  < G G56IL7rMac1

CultureinS17stroma

8 6=>=QG 5 = 5 L L : R 8 6 == Q G5=5LL

MitomycinCtreatment

857

 5jY=R6^L_LjY=>=  A6  6:8  R==

MTTassay

 5 jY = ^R^5R<RR_R5R< G__  A6  8  -jY=8H^S6_  5GG

PARTIII

RESULTS

Results

1. Role of HDAC7 during the reprogramming of preB cells into macrophages

/565 @ RR 6 RRG 6L>:5 G R R@ Q ^9 -_ ^5 _ R 6 6 R ^/59Y _ =86R6L>:RX9: ^6_=86R6L>:R6^6_ R5/566R R R =5 /56_ R5/56 =>96 @ ^R5_

Figure1R:ModelforreprogrammingBcellsintomacrophagelikecells6 R^66_

Results

6 RG 6L>:5 R@Q^5 _

1.1. AnalysisofclassIIaHDACexpressionduringreprogrammingofpreBcellsinto macrophages

95 -R_:6- < 6 /56 G^/5655_R G/56RR R ^9 -_ 5 /565 6 ^5_5RRag1Pax5G R 5 @ c/Ebp Csfr15 5GR

Results

Figure2R:HDAC7expressionisdownregulatedduringthetransdifferentiationofpreB cells into macrophages. -R_:6- 6 6 /565^^R_R6 ^U__

R /56 G G6565 `5-H85>5:555G R 5 =>96-<R ^9-_</56-<5 -5^9--_

Figure 3R: HDAC7 and B cell transcription factors are down regulated during transdifferentiation of preB cells into macrophages.  G /56 ^ ^R_ R R 6 ^U_ _

Results

5</566 R6^5 _  G /56 -<5 6 G:5^9 -_

Figure 4R: HDAC7 is downregulated during transdifferentiation of preB

cells into macrophages. ` ^ 56 6 _ / :6 6 R 6 6L>: 6 R/59Y5 R 5 G -5

Results

8/59YRG/R56 6  @ ^ 5 _ 66/56R G5:55/566 ^9-_665 /56 R R ^9-_55G /56GR5/56R6 H /56 G GRR 5

Figure5R:RTqPCRexperimentsforgeneexpressionchangesofHdac7,Pax5, Rag1andCsf1rgenesduringreprogramming.-R_:6- ^_ RR R 6L>: 6 :

Results

1.2. Study of the effect of HDAC7 reexpression in C10 cells and its possible interferencewiththeacquisitionofthemacrophagegenetranscriptionprogram duringcellularreprogramming

R/56 R5RR 66/56R9 6 ^6R/56 _ 5 5 ^6R=86 _ 5 R 5 6 /56 R 56/56R9 ^9-_

Figure 6R: Downregulation of endogenousHDAC7proteinlevels during reprogramming.  G 6 6 /56 6 R

/56 R6 5 R 6 = 6 G 6R=86 6R/56 R R R

Results

6R=86RR55 5 ^5 _ -@G5 R6 /56^6R/56_G@R 55GG, R5 5R ^5 _ G R6 /56 5 ^XH_ :566=95`>XX 5X^:XX^9-_X 66=9 ^9-_

Results

Figure7R:HDAC7reexpressioninterfereswiththegenetranscriptionalprogramofthe converted macrophages. /R ^ R \ _ XH : `>XX : R GR6/56R - 5 6

5< G R6 /56 5^Cxcl10,Ccl9,Tlr4Il18_ 5 ^

Results

Fcgr1,C3Mbp_R ^Tlr3,Tlr9Ccr5_^9-_

Figure 8R: Gene ontology analysis corresponding to Cellular Components and Molecular Functions. /R ^ R \ _ XH 6 6 = 9 R G R6 /56 R - 56

Results

Figure 9R. HDAC7 re expressioninterfereswiththe gene transcriptional program of the converted macrophages. /R 6 XH G 6 6

6G_:6-5 GR5 Cxcl10, Fcgr1, Tlr3 C35 G@ /56 5GR 5:5^9-_ /56_ /56G R

Results

Results

Figure10R:RTqPCRformicroarrayvalidationofselectedgenes.-R_:6-6 6 R 5 665 95 5 5 5 5 65 G /56 R 5 /56 R6 R :6 ^ ^8>=_ 6 5 R 5

5 et al., R 5 GR^5_H RG/565 G@ R ^9-_

Results

Figure11R:HDAC7reexpressioninterfereswiththegenetranscriptionalprogramof the converted macrophages. / ^9- R _ _ XH : R GR6/56R - 5 6

1.3. AnalysisofthephysiologicalfunctionofHDAC7inpreBcells

</56R5

Results

RR 6  -<5 ^-<5_ @@ /56 /59Y R 8@5 @@ /56 6 5Itgam,Fcgr1Ccl35G^9-_ /56 R R

Figure12R:HDAC7knockdownleadstothederepressionofmacrophagegenes./59Y -<5-<5 /56 /5 5 9 6 -<5 G-R_:6-6

Results

1.4. StudyoftheinteractionofHDAC7withparticularsequencespecifictranscription factorsinpreBcells

95/56R 5 ^:5 6 =5_/56 R56 /56  /56 =>96 RR5 G G R 5>55`5-H8:5^9-_58R8etal. =>96 Y=::G=>96R ^8R85_

Figure 13R: HDAC7 binds to the transcription factor MEF2C. 6 6 G /56 =>96 R/56G `5-H85:5>5

5 G G=>96Y=::R G@ G /56

Results

6  R /56R =>968R855 5GR_^\R_^Y _ 5 ^9 -_ 5 ^Fcgr15Fcgr2bFcgr3_G/56=>96 /56G=>96 GY=::R

Figure 14R: HDAC7 interacts with MEF2C and alters the upregulation of several genes.  G R =>96 Y=::5 R R G/56G

6^6:_6 /56 R R /565@G

Results

9 =>9 G 5Fcgr1,Cxcl10Itgam 5 G Fcgr1,Cxcl10Itgam6 RG=6:G /56=>96_:6- /56=>96G=>9G5 R/56G6R ^9-_5R5/56 =>9

Figure 15R: HDAC7 binds to the transcription factor MEF2C and is recruited to the promoter of key genes for macrophage function in preB cells. 6 6 /56 =>96 =>9 G Cxcl105 Fcgr1 Itgam R

Results

- 6

1.5. Analysis of the effects of HDAC7 reexpression on the functional properties of thereprogrammedmacrophages

5Getal.^_ R 5 @ =R ^Itgam_ R 5 R @ 6 G R -@G5 Y:8 G R @ ^5 _ < /56 R 65 6 G 6R=86 6R/56  <=R66GG R R =R R 6 G 6R=86 6R/56 /5 ^=9_ 6 /56 G@ =R 6 < G665 6R=86^9-_

Results

Figure 16R: HDAC7 reexpression interferes with the functional properties of the convertedmacrophages.:=9=R6R6R =866R/56R

>66/56 R=R6^9-_ /56 G 6=R<65 6/56R6 5 R G RR 6R/56 G RR 6R=86^9-_

Results

Figure17R:HDAC7reexpressioninterfereswithMac1expressionandthephagocytic activity of the converted macrophages. = 6 R 5 6R=86 6R/56 R G

956R6/56 Y:85G@5G/56 G@Y:8GG6 R @ Tnf5 Il1 Il6 ^9 -_ 5 6 /56

Results

Figure 18R: HDAC7 reexpression interferes with the inflammatory response of the convertedmacrophages.>/566Y:8RTnfa,Il1bIl 666R=866R/56R 5 G Y:8 -<5 <G-R_:6-

<65 5 G 566/566 =R6RR6 /56R9=R6G GR6/565=RG R ^9 -_ /5 6-55 6^9-_ 5 /56 R R /56 5

Results

Figure 19R: HDAC7 re expression decreases Mac1 protein levels in the reprogrammed macrophages. / =R /566

Figure20R:HDAC7reexpressioninterfereswithfunctionalpropertiesofreprogrammed macrophages. -R_:6- 6 6 /5 5 65 9 -5R=86 -5R/56 5

Results

G

1.6. Elucidation ofthecontribution of the interaction ofHDAC7withME2C and its catalyticactivityontherepressionofMac1

5 5 /56 =>9R 5 G6 ^5 6 5 6 Y5 6 `5 6 _5 6 /5 _ G G/56=>965/56 =R56/56X9: @85 =>9G^/56R@=>9_59 ^/56RR_5 ^/56RR _5=>9G^/56R`5`5_ ^/56R`5`5 /56R/5_  6 < 6 =R R 565 6/56 R=RR/5 6/56 /56R=>9 /56 G =R R^9-_

Results

Figure21R:InteractionofHDAC7withMEF2Canditscatalyticactivityareessential for repression of Mac1. : =RR ^_ -R_:6- 66^_6=86R X9:5 =86RX9:R/565 =86RX9:R/56^=>9_5 =86RX9:R/56^`5L`5_5 =86RX9:R/56^/5_5 =86RX9:R/56^R_ =86RX9:R/56^R_ R^8>= G6 \

2. RoleofHDAC7inBlymphocytedevelopment

2.1.HDAC7expressionpatterninthemurinehematopoieticsystem

/56 G5 < 65 X := G ^_ ^7LLL_ G = G R6

Results

G<G /56G6G ^9 -_ G /56 RR

Figure 22R: HDAC7 is highly expressed in lymphoid cells but not in the myeloid lineage. G - := G ^_ 6 /56 < ^7LLL_

Results

2.2.GenerationofHDAC7conditionalknockoutmiceinBcellprogenitors

/56_ Ginvivo,X6/56 /56R5 /56 invivo, /56 @@ G =^65_ 5 /56 5 /56^R_ 5G><H^658585_5 G6R5

Results

Figure 23R: Strategy for generating a conditional knockout mouse for the specific deletion of HDAC7 in the hematopoietic system. = G 6: @ /56 ^/566:L@_66G Mb156RR

2.3.AnalysisofthehematopoieticcellpopulationprofileinthebonemarrowofHDAC7 knockoutmice

</56 G/566:L@6=GR6@LU 5/56UL@6=GR6@LU^=_5^8:Y_5 ^/_G^:_R@R<G G/56@@^U_ ^9-_

Figure24R:BcellpopulationisdramaticallyreducedinHDAC7knockoutmice.9 ^ _ G G ^ _UG/56@@

Results

<65 6 /56 G  G _ R ^U6U=R_ R ^U6R=R_ 5 RRG /56@@^9-R5_65GGR R5^9-R _

Figure25R:ImpairmentofBcelldevelopmentattheproBcellstage.9 ^_GG^_R R

-@G5 ^U6R=UR_ ^U6R=UU_G /566:L@6=GR6@LU /56UL@6=GR6@LU ^9 -_ 5 /56 G

Results

Figure26R:ImmatureandmatureBcellpopulationswereabsentinthebonemarrow of HDAC7loxP/ko mice. 9 ^ _ GG^_

<65G/56 5 ^XU6GU_5 ^XR6GU_ ^U_ G/56@@^9-_65 /56G@

Results

Figure 27R: HDAC7 has no effect on myeloid development. 9 ^_GG^_5 G

2.4.Analysisofthehematopoieticcellpopulationprofileinperipherallymphoidorgans ofHDAC7knockoutmice

/56 _ 5 < ^ G_ /56R /566:L@6=GR6@LU /56UL@6=GR6@LU  /56@@^U_ ^9-_8@5U 6G/56R^9-_

Results

Figure28R:BcellpopulationisreducedinthespleenofHDAC7knockoutmice9 ^ _ G G ^ _U/56@@

Figure 29R: B cell population is much smaller in the peripheral blood of HDAC7 knockout mice. 9 ^ _ GG^_UG/56 @@

<65 < /566:L@6=GR6@LU @@ 5

Results

< ^U6R=UR_5 < ^=_ ^U6U6Y6R_5^9H_^U6Y6U6R_ ^_ ^U6Y6U6U_  G G G/56R5G G/56@@ /56^9-_

Figure 30R: Marginal zone, follicular and transitional B cell populations were dramatically reduced in the bone marrow of HDAC7loxP/ko mice. 9 ^ _ G G ^ _ <5

/56@@ ^9-_5 G /56 R 55 G G^9-_

Results

Figure31R:HDAC7hasnoeffectinTandmyeloiddevelopmentintheperiphery.9 ^_GG^_

2.5.HDAC7knockoutmicehavesmallandunstructuredspleens.

XG@/566 < G -@G5 /566:L@6=GR6@LU@@_ ^9-R5_<65/566:L@6=GR6@LU /56UL@6=GR6@LU6 G ^6 _ G ^ @_ 8@5 G /56 5 ^9 -R_ 5

Results

5/56@@ =5/56 ^9-R6_

Figure32R:HDAC7deficietmicehavesmallandunstructuredspleens.A.8 /56R@@ _ B. /6 ^ _ ^ @_ /56R 5 C and D. /56@@ 5

Results

2.6.GeneexpressionprofilingofHDAC7deficientproBcells

<65 6 /56R 5GR6R Hdac76L@6 GR6@LU Hdac7UL@6 GR6@LU = 6 /56R R55RR 5R55 R5R55R 5G_^G-_H G/56 R 5 6 R @R 8@5 G G /56 R R @5Itgam,Itgax,Ifi204,Tet2,Ccl3, Ccl4 Ccr25 ^G -_  G R 5 @ ^Fosb,Egr1,Crebzf,Cebpb,CebpdCebpz_

Transcription Factors Jund 1,0

Gene Log2 Fold C/ebp Cebpb 0,7 change Family Ap-1 FosB 4,6 Cebpz 0,6 Family Cebpd 0,7 Fos 3,2

Jun 2,7 Runx Runx1 0,5 Jun B 1,8 Family

Results

Runx2 1,2 Chromatin-associated Factors

Runx3 0,7 Gene Log2 Fold change

Brd1 0,65 Notch Notch2 0,7 Family Brd8 1,0

Notch3 0,7 Cbx4 0,6

Chd2 1,0

Klf Family Klf2 1,6 Chd6 0,7

Klf4 2,0 Chd8 1,0

Klf6 2,18 Ezh1 0,6

Klf7 0,84 Kdm2a 1,1

Kdm5a 0,6

Zeb Family Zeb1 1,7 Kdm6a 1,66

Zeb2 1,3 Kdm6b 1,8

Mll1 0,51

Others Foxp1 1,4 Mll2 1,0

Egr1 3,5 Mll3 0,6

Nfat5 0,6 Mll5 0,72

Mef2a 0,7 Mllt10 0,6

Crebzf 1,0 Mllt3 0,55

Stat3 1,6 Setd2 0,74

Ikzf2 0,5 Setd3 1,0

Setd6 0,6

Setd8 1,0

Results

Smarca2 0,71 Cd33 0,92

Smarce1 1,22 Itgam 1,3

Suv39h2 0,56 Itgax 0,7

Suv420h1 0,8 Fcgr4 0,8

Tet2 1,0 Csf2ra 1,4

Jhdm1d 1,0 Csf3r 0,8

Jmjd1c 1,52 Cr1l 0,7

Jmjd6 0,8

Hdac9 0,82

Mbd1 0,92

Mbd6 0,82

Hdac9 0,82

Inflamatory-response genes

Gene Log2 Fold change Myeloid-related genes Il10ra 0,75 Gene Log2 Fold change Il12rb2 0,85

Ccl3 1,2 Il17rb 0,56

Ccl4 1,2 Il1f9 0,76

Ccr2 0,63 Il1r2 1,1

Ccr6 0,52 Il1rn 0,55

Ccr9 1,2 Ifi203 0,55

Ccrl1 0,52 Ifi204 1,13

Ccrl2 0,6 Ifng 1,6

Results

Tnfaip2 1,0 Mapk8ip3 0,80

Tnfaip3 1,4

Nfkb Family Nfkbia 0,92

Nfkbiz 2,3

T cell-related genes Nfrkb 0,66

Gene Log2 Fold Ikbkb 0,72 change Ikbkg 0,7 Cd28 0,77 Pkc family Prkca 0,7 Nfat5 0,85 Prkcb 0,7 Lck 0,8 Prkcc 0,9 Sla2 0,84 Prkcd 0,8

Prkd3 0,6 Intracellular signal cascade

Gene Log2 Fold change Protein Ubiquitination

Mapk Map3k2 0,85 Gene Log2 pathway Fold change Map3k8 1,34 Cblb 1,2 Map4k4 0,95 Birc3 1,2 Map4k5 1,0 Birc6 1,8 Mapk14 0,6 Ube2j2 1,3 Mapk1ip1 0,72 Ube2b 1,4 Mapk1ip1l 1,6 Ubr2 0,6 Mapk6 1,6

Results

Ube2i 0,8 Cell cycle related genes

Mdm2 1,4 Gene Log2 Fold Malt1 1,1 change Cbl 0,8 Bcl2 1,1 Cnot1 0,93 E2f5 1,0 Cnot2 0,8 Gadd45a 1,39 Cnot4 1,0 Gadd45b 0,51 Cnot7 0,83 Gadd45g 0,7

Pten 1,0

Trp53 1,1

Table1R:Changesingeneexpressionof Cdc40 0,5 importantmyeloidgenesintheabsence Cdc42 1,23 of HDAC7. = Cdk13 0,7 R /56R R 5 Cdk16 0,73

R Ccnc 0,62 Ccnd1 0,9 5 R 5 Ccni 0,7 R 5 5 R 5 Ccnl2 0,61 Ccnt2 0,74 G_

Results

5 /56 R 5 Cd28, Cd69, Il Lck5 R Runx1,Runx3Nfat5

< G /56G^XH_ :5`>XX^9-_

Figure 33R: Changes in the regulatory network of B cells and macrophages in the absenceofHDAC7./R^R\_

Resulta

XH:=9R G/56

G G /56 G XH 5 5 @ 5 5 5 `>XX =5:`5 5 <5 G_R 5 5 @ 9 R 66^Fosb5Jund5 Junb5Crebzf, Klf4,Egr1, Cebp, Cebpz, Tet2, Itgam, Ccl3, Sykb, Map3k8, Nfkbid, Nfkbiz, Cd28, Usp7 and Mdm2_ G_-R:6-^9-_5/56 R

6GR/56RR RG@G/56 R 6 ^QR/56RR_5Fos, Jun, Cebpd, Ccr2, Ccrl1, Ccrl2, Cd33, Fcgr4, Tlr75 Mapk3k8 Gadd45b ^9 -_ G /56

Resulta

Resulta

Figure34R:ValidationbyqPCRofselectedgeneswhoseexpressionwerealteredin the absence of HDAC7. 6 6 ^Crebzf, Klf4, Fosb, Egr1,Cebp,Tet2,Cd28andMdm2_G_-R:6-

Figure 35R: HDAC7 deficiency leads to the upregulation of macrophageassociated genes in proB cells.  G R /56@@RRR RRRG/56 G

2.7. MolecularmechanismsunderlyingHDAC7repressioninproBlymphocytes

5 G5 /56 6 GR5G _ ^:5 6 =5 _  /56 =>96R /56 =>96RR5 5

Resulta

^6:_6959G Itgam5Fosb,Cd69,Cd28,Ccl3,Mdm2Egr1=>9G =>9 G.6R G=6:G=>96/56_-R:6- Itgam5 Cd69, Cd28,Ccl3,Mdm2Egr1=>96/56G =>9GR ^9-_@5/56 R=>965

Figure 36: HDAC7 binds to the transcription factor MEF2C and is recruited to the promoterofnonlymphoidgenesinBcells.66

Resulta

/56=>96=>9G Itgam5 Cd69, Cd28, Ccl3, Mdm2 Egr1 - 6

2.8.AnalysisoftheresponsivenessofHDAC7deficientproBcellsexvivo / R G@ G /565 R /56 @@ G YR-5 ^65 _ 5 R /566:L@6 GR6@LU 6 YR-5 R 6YR-^9-RY_=^R^5R<RR _R5R< G__ /56R R exvivo5^9-R- _

Figure37R:ProBcellslackingHDAC7failtoupregulateIL7Rexpressionand,thusto proliferateinexvivocultures.U=G@@R

Resulta

/56R R YR8G/56 R5 5 6G 6 @ ex vivo G5 R G/56UL@6GR6@LU/566:L@6GR6@LU 8GYR 8@5 G R @ /56 6 @=R6^9-_

5 5 G /565 G@R:R/566:L@6GR6@LU R 6 @

Figure 38R: ProB cells lackingHDAC7expressMac 1. = 6 6 @

PARTIV

DISCUSSION,CONCLUSIONANDREFERENCES

Discussion

Discussion

 5 5 G G /5 _ > ?  @ R /5 G _ =  GGR/56 G6/56GG@ 7 5 R 656 5 6 G R G=invitro6 in vivo 6 @@

1. RoleofHDAC7duringthereprogrammingofpreBcellsinto macrophages

G@ G GG < G

Discussion ^656:56H5_ G^5_ /56 6 G 5 = 5 /56 R @ 95 /56 _ @ GR6=R 5 G G_6G6= R @ 5 5 /56 =>96R=>9G =>965/56 H/566 R 5 G ^9_

9 5 G /565 /565R6 ^=56 :5_5=>965=>9 G5 G 6 5 ^85 _=5G=>96 6- =5:` ^`5 6 @5 _=>966^X@5 68R85_=58R8etal. =>96 5 GG ^Y=::_G=>96Y=::R

Discussion

^8R85 _ 5 =>96 5G5 ^X@5_ R/56R G 8R8 ^_5 5GGR_5^\ R_^Y_55 ^ Fcgr15 Fcgr2b Fcgr3_ G /56 =>96 =>96 /56 G Yetal.6:R8_G >55 >9 9HH R /56 G5G /56 G^Y5_

H5R5/56=>96 =>95 =5 =>9 /56 G R  /56 G R R 5

Discussion

Figure1D:ModelforHDAC7mediatedtranscriptionalrepressioninpreBcells. /566RR5/56 =>96 R /56 R 6 R

95 6 /56 5 /56 5 /56 R R5 G6 <56^6Y_5/56 R G G_RG^_56_56:5

Discussion

6 `5 6 `5 6 <5 _ 5 /56 < 5 6 G ^_ ^_ /56-5/56G 5 ^_56_56:56`5_Y55 /566Y5G 6@<^<5 _55 /56 R GG 5/56< _ 5 GG5 5/56G6H /56 6 R R /55 G /56 R 5 /566R^_5GGG ^6-_/5GG /566- GR 5 5 5 /56 R 5GRG GR

Discussion

2. RoleofHDAC7inBcelldevelopment

G 5@: 6RR G 5 5 ^6G5 6 <5 6 8R85 _ /5 in vivo /56G@5 /565 G5/56=>965 RR5 R 5 /56 =>9 G R5565656 >G=>96Y=::^8R85 _ 5 5 @@ 5 =>96 /56R^ 5 5 G_555 R_G5G@ ^9_=56/56 R R@/565GRR R 5 R G XH 5 G /565 R G 5=5 5G G /56 5 5

Discussion

5 G_5 G/56 GR/56RR R G@ G /56 5 /56R R R 6@=Rexvivo 5 G /56 R

 G /56 6 5 :55 >55 > `5-H8 G /56 G@ 5 /56G R 56:55 5 5:5:;^6G5_ 6:R8_5-RRetal./56:5 R5:5/566^-RR5 _ G :5 R G :5 R H @ :5R  :5 6 5 /565 G 5G_

Discussion

5 /56 R

Figure 2D: Model for HDAC7mediated transcriptional repression of non lymphoidgenesinproBcells./56=>96 R5G@

Conclusion

Coclusion

1. RoleofHDAC7duringthereprogrammingofpreBcellsinto macrophages.

/566^R_G 5

R /56 6R

/56R6R@ R

/56 R6

/56 @@ R R @ 5Itgam,Fcgr1Ccl3

RR5 /56 =>96 =>9

=>9/56

2. RoleofHDAC7inBcelldevelopment.

/56RG@R

9 5 /56 =>9 R

Coclusion

/56RR 5

G/566R 5:55>55>`5-H8

/56R6YR-

/56R6=R @exvivo

References

References

A

555=B5-5<R5<5/_555YG5`556555 5 Y5 5 H5 5 Y55 55 `5 8@5 >5 8@5 5 85 =5 G58>^_9UR@R @ G  6 7R

5G5 5 G?@5 5 85 5 5 >5 8=G5 5 Y5 =5 ‹5 /5 85=^_>R9->6R: R=67R

5@G5 :5 -G5 =5 Y5 5 85 = ^_  ^GGR_R=6 7R

5@G5:585=^_> @7R

55 <5 55 5 65 >5 /5 :H ^_ 5 RR=G7R

55X59@5 -5=@55>^_ 5 G-:<58857T

5G5 =5 /5 5 =5 Y5 -5 X ^_ :R <-877

5@5 =5 -=5 5/5 X5 /<5 65 85 `G5 : ^_ >/^_7</56? 5R

5G@5 /5 =<@5 :X ^_  G R =G7R

References

55=5`556<G5=:5:55=555‹5585=5-5 55RG5-5H5><^_=>96 G67R B

@5 5 @5 5 <<5 85 =`5 55 H5 >< ^_ / _ LR @ G <=67R 5585885=5Y5:5^_:/56 -<5GGG66^^__ R5 85=5 55`5 655 5/5 5?5 =5@5 8555`*58595558G59^_>/566 ?<5T 5 X5 =5 >65 <5 5 55 5 `G@5 5=5 G5 65 `5 5 85=859555-5=5@5=5/-5555=5 6 ^_ >5 _ G67R @5->55=H^_6_/56 67R 5:5:5=5555=59@55X595-5`5`5 855855=5:585>55`^_G @7G @G67R =5 55 <5 5 @G5 /5 G5 8>5 55 ^_ X 7R 5X5=555H585565=/5-@5-6^_>:R @G6 H7R

References

5X55-5`5=5H56565556555Y55 -`5 >5 -= ^_ R <9R@ R7R R55:5=^_/=H7 R @5``5<5558/5/<5`5=58^_- 5:R^5:>L-R_>=H 7R

@5 85 `5 8`5 5 X5 6@5 =5 5 55 5 85 5 55 Y@5 `5 5 Y5 R85 5-5 5 6>5 `5 95 5 =`5 5 5 5 995 85 5 5 95 /5 >5 `G5 5 /G58^_>/ G?667R

5X5X5Y^_85- 7R 5 Y5 X<<5 5 Y5 Y5 5 @5 Y55 =5 5 <M<5X556^_GR65GRR 67R 5 >5 :5 = 5 5 -  ^_ 5 G<-7T 5 5 5 : ^_  @ <7R 5Y5555<585Y5:55:5X5-^_>_ ^_G>6=7R 5 85 =5 85 -<<5 X5 :5 95 5 -5 95 8 ^_  < =>9 GR 687R

References

R:X5Y5_55^_X6 7R 555`5>5>G56958G5585/^_:5R -95G6 >:X7R 55555‹585/^_5G_:R:^-9R_ 67G>=H7R 5`5Y5`65:@5655/5>55@5=^_>6 X55RL RX7R 5555-55`G5-5>5X5-585556^_  57 X @ 67R 5585=^_8RR 6H7R 5 = ^_   5 - 7R 5 Y /5 8G5 55  =5 :5 55 Y5 G5 865 :5 =5 55-595-5X5^_5G 6867R C

65X5=5Y5X555658555Y55:5:5 Y585>595>5=5>5:5=585X5:5>=5X5 :5X5558@*5655Y5:56585=>5855X5 5 ^_ / 6R-><^`6_ G_ /X<67R

References

65 5 95 5X ^_ > <8=7R 65 X `5 85 5 85 95 :=5 85 5 95 5 5 5 =-5`@5=5<5565YG5Y`585^_5`5:RYGG< =67R 6585=`55556Y5-555/555H5><^_/ G =67R 65 85 <<5 85 Y5 85 H5 >< ^_ 5 6 :<58857 R 65 85 5 5 Y5 85 ‹5 5 -5 55 H5 >< ^_ / G66 7R 6=5`55X555@55Y5:5=X5G585 `585-=@585=@55-58=555/55RG5-5H5 ><5 Y5 5 /5 8=5 G5 <Y ^_ / 67R 65 5 =5 >=5 5 =5 5 ‹5 65 >5 /5 8=5 65 9Y5 5  ^_  -<5R -<5R @ <X7R 65Y959555>5X56^_<9R@ GG87R 65 5 @5 ^_  @ `*R@=67R

65 85 H5 X65 =@@5 /5 >5 ^_ >/56 ?:Y8H<>55 65858RX5<5=6@5:^_/G7 GG=6X7

References

65 5 `5 =5 5 85 5 5 :@5 `5 Y5 65 Y5 /5 85 -/ ^_  8L8<9R@5967R

6565`565X595<5=Y5-5=5565H55 =5=^_>Y6R= ?87R

6555956556^_6/567 G6==7R 6555X5>5585=R559556556^_ 6/56=>9R >-UG958>7R 6G5658G5555555=^_:67 <7R 6G5655=^_6H 655>5859=565-5655H5=5@55- ^_@R >=H7R 65 8 55 -5 -5 :5 6 ^_ = YR G 8H68 7R 6=5:5H5><^_=RG =>9R R  = 6 7 R D

5 5 95  ^_  - G 6 / ^/56_R8H7R

References

5 65 Y5 =85 /5 `65 5 ` ^_ >6:LR /?<7T

5 95 X5 =5 6R=5 5 8<5 -5 X5 =: ^_ 6LR @ G5/56 67R 5 -5 G5 /5 Y5 5 ^_ >6 <5 GGG67R XG5  X5 XR<_5 <5 -5 5- ^_ - R G-<5-7R 5=55=5<@5-55>555=5=585`5`5 =5 `5 5 6@5 5 65 `>5 5 >5 5 65 5 <5 >5 85 95Y5X55/5`5/5=5@55565`5=5 `5 5 YG5 :=5 =R5 Y5 =5 X-5 <<@5 <5 :5 =5 85/585=58<@55@@5`5555:5:565XR `G5X5655`595@5YX5/55`<5585 `^_/566Y <7R G55-5`=5:5:5555/^_R = R 6 @ R 7R `5-585/^_-Y= GX>6:87R 5 55 8G5 55 85 ‹5 5G5 `5 :5 5 5 = ^_ XG:6G 7R 5/55X5H5855@58555`6565/5:5 -55 -@5 -6 ^_  96 <7R

References

:5 -5 Y5 >5 95 Y5 `5 <>5 5 X5 55 9 ^_ 86 ^_ /565 R 5 < 6-R7R _595Y55Y5>5558555-5 <5 -5 `5 - ^_ : G @ R < 6  >6 = 7R _5 95 =5 =5  5 5 5 5 Y5 >5 =5 <5 /5 =95 5 =5 <5 65 /5 =65 -5 =/5 :R5 /5 85 5 `5 - ^_ < :R @ >=` 6R5` <  = 6 7 R _595`5/5955`555=5/5X55> ^_ /565 R 5 < 6-R7R 5 55  =5 :5 8G5 55 65 X5 =B5 -5 X5 ^_ 6655L G ^6L>:^__R R:<588 57R 5 65 <G5 6H5 `5 5 X5 5 85 X`5 <5 66 ^_ 6 G  5 - :67R E

>G56585/58G5^_:5-9G5R 5:RX7R

References

F

9559<<5=5:55555=55XG5=5:5-5 5565<R5Y5558G56565X5:585R95 -5 85 / ^_ 5 G ^==_ 6 ==L-9G55 7R 9555655=55<5X585=5X5‹595 5X5@5``5=5855X^_5:>L-R:><6 G>R9--7R 9<55`5=8585=5‹55/555-555RG5-5H5 ><^_-_@: <58857R 95=855-56555-5=5-@5-55=@5:555-5 :5<:^_8G85 85/5G<7R 95565-5555G595585655-55<55-5 5=5 :5 65 ^_  < 5/5667R 955_59595=5/<5=5555>^_/ /56/5667 R 955_595/<5=5X5=X5Y<5=55555> ^_></56 6/568=-L5 /5 -5 :5 5 ?5 H55 :5 -X ^_ 5 =>``R R=67R

References

9555`55^_8_6 -67T 9=5<55Y5X5655@<555=5=-55=55:5^_ =5 =RL<- 6 67R 9@5555H@585/5=5=@55G5`5H@G585`@5/5 /5855=5=@5<5@5^_ >6=7R G

X5 Y5 65 = 55 5G5 95 5=5 : ^_ 6 </565G 6R X5Y5Y5-5<5>5-G56<^_- G=RG_<5-7R X5 Y5 65 Y5 5 =5 <5 -5 ^_ RR 56 X6G7R X@5 65 -5 `>5 X5 Y=5 /5 /5 95 -5 `5 8`5 =R-<5 >5 RG5 -5 H5 >5 `5 55 55 85 H@5 8/5:5=5=@@5/`^_=6R8L @R7R X5 `5 =5 5 5 ^_ @5 R  8 7 R X5`5G5=55/5=5555:5585855^_ @_67R

References

X5`585=55958555-X558`^_X 6^/56_G7 G/56G=5=6 7R X<@5=5585<55585>^_5 RX7R X595=5`=^_=<57/5/56@@ =-78R^_R X5 Y5 `5 5 8G5 >55 H5 >< ^_ 5 R RG < R=67R X?555Y/^_/7 <-<7R X?555Y/^_/56G5 -:67R XX5855^_5 =>9=67R XX5855Y5<55555=5Y555585>55^_ /= 67R

X5 5 Y5 =5 X5 / ^_ >= 7?=7R

X<5 6 =5 8G5 8Y ^_ - GRRR<:<5885 7R X5 5 -5 5 : ^_ = < /5667R

References

XV5 5 5 85 <<@5 -5 =5 >5 :5 85 Y5 >5 X5 - ^_  >G R 5 5 X7R H

/5 6 /5 `5 5 5 5 5 5 5 85 5 65 5 X ^_ :`5 65 G:<5885 7R /G5 =5 =5 -Y5 H5 >< ^_  7?< -X7R

/@555Y585`555555@5=5<555855 58V565:5H585=565X5=5:5Y5/5`G55 856^_>5/G/G G ? ^ _

/5 85 `5 `=5 5 65 X<5 6=5 8G5 8Y ^_ > G/56RG?: <588555T

/5855Y/^_:G/56R <GY=7R /55-3<55Y@5`^_5 G  6 =G 7R /5 `5 >5 :5 Y5 `5 `5 X85 @5 65 85 8 ^_  Y9RG5R-<5 @=67R

/5 65 Y5 85 5 = ^_ >8 R65?:<58857T

References

/5 55 /5 5 5 5 Y5 H5 65 Y ^_ = GR=7R /5 5 =@@5 85 85 :5 65 5 5 65 5 =5 65 =:585>565:595-5Y5655555-^_  67R /565`5=85-=5>5:5695:5:55-5=5555 -555RG5-5H5><5=`55^_- G@=67R /5:H^_G_55RR 6:7R /5 6 55 5 `5 95 65 H5 5 XG5 :X5 55 >5 8G5 8Y ^_ 5 /56R :<58857R /@5 95 5 =5 H<5 5 5 55 :@5 =Y5 85 / ^_ 9X55RGY@7R /<5<5/5X65/@5-5`5:58@555/5Y9555Y5Y`5 85-`5656565`5555Y5/555X5-5 YG@5 5 85 -5 5 55 65 X>5 `5 =5 -5 ^_ / G RR 6<7R /<5<585-`5/5X59556565/@5-55YH5 65 85 ‹5 65 65 `55 5 5 5 5 X5 -5 65 X>5 -5 ^_  <X7R /<58556585<55>5Y565Y5:5655=56585 /5 X5 6` ^_ 8 G R R_=67 R

References

/G@5>55858555/5<5:5:5-5-5=^_ GR:<5885 7R /@5 85 5 5 6<5 5 95 5 85 H5 55 =Y5 R @@5 5 ^_ RY :6 R 5G>66-7 R /@55YV55`55955X5<5-G595Y55`G5 `:5XG5Y55955=95@5-=5=5/65H<5 `5/@5^_@R@ 68:67R /5 =5 8G5 55 5 = ^_ :6L85: 7R /5 5 `5 55 /5 -5 /5  ^_ - 9HH G 8- =H68:Y8H7 /5 >5 65 5 9@5 5 5 5 `@@5 -5 5 X95 5 `5 85 6=5 Y5 -5 @5 ^_ 6 < 67R /5 5 6@5 85 @5 Y ^_ - R@ G G@ G-7R /565/<5`5=@55^_YRR GG<7R /5 > 5 5 5 =@5 >55 X5 =X5 `<5 5 Y<5 =5 ^_ < 8R X7R

References

/55 =5-5 X55>=@G55585=5 655>5 =55 ^_GGR <7R /GG565X55585-5`55555<65555=55 955:^_/56GR<7R /5 5 5 ^_  G 6==7R I

5/5X5865@558@5<5`5:^_ -5X @ G  7R @5 5 @5 `5 5 5 X5 `Y ^_ :>< :6-7R 55-5656565-55:5=58595-55585 -5X5=5955:58@565X5:5=55565=65 X56^_<6 5 6-7R 5 5 X5 Y 5 < ^_ 6 R 7 <7T 565<@5/55^_>G R< -67R 5<55=5/5=55H5<@58^_-YR `Y XL8  7R

References

J

5>5Y5`565<85Y5=/5`5/<55/5-5/=56555=5 <5 <5 H5 65 Y5 `85 Y5 /5 5 86 ^_  R -667R 55<5‹565555<@555:55Y55Y5 5/5Y5/-555=*5=5/595658^_6 G/5G_=8=7R 5`5/55856=5:G5=58@@55555585/ ^_ - G R G G -9R@R7R 5Y585565=/55=585>5Y58556^_/ RR67 R K

`5>5-35X5^_6>:R R=67R `<55`55X5855:55-55/5>5=5=5-V5Y5 85 >5 55 /X5 5 5=5 5 85 G5 6 ^_ 6 G /56 G>=H 7R `5/5555566585655/5`G58^_= 67R `5 / 5 5 =5 H5 :5 >5 -= ^_  8=-R X7R `5/5555566585655/5`G58^_= 67R

References

`5 / 5 Y5 6/5 `5 55 /5 665 >5 -= ^_  </56567 R `5=55555:@5/55Y58@5-595-85=555:@5 X585>565^_9: <9R+67 R `5 /5 5 5 =5 5 Y5 /5 65 5=5 H5 ><5 5 > ^_ /6-6L6GR 7R `5/X55>^_/@ G=67R `5 55 X5 -5 /G5  ^_ ‹=RR G @_7R

`5 Y5 =5 >5 55 585 5 =5 G5 >5 Y5 =5=65 -55H5`6565Y565/565`9^_>8-@/ <9R+R6?67T

`5/5Y5/5558558555/`^_H6 /56 G G 67R

`5-5658=^_>^/56_ 76R?87R

`5H595855585Y5565:<<5955=5`5 5 Y 5 < ^_ /- G@ /56 GRG:<58857R `5HY5<^_/56GG7 67T

References

`5 5 65 X5 8<5 5 @5 Y ^_ 5 =5:`R=>96 R:<58857R `<5 8 ^_  7 < 7R `5-5W55<5-565=:5>=585856555>5 85@5=^_:RR<9R@R G67R `@5`5Y555/56Y5`5=^_YR R>9>6=7 R `5=5=565595=5/5`=5/585855-G5X ^_G6 5<<7R `55858555@555`55/5>55855558555 @5 5 `5 -5 X5 ` ^_ @ <5RG 67R `5H58585:555X555=*585=@5>5=5=5/R5 55`<55=595=55^_8=H>-: /56>=H7R `55658565X585‹5Y5=5=55Y55-=@5`5R 95-^_-9R G<7R `5 =5 5 Y5 5@5 ` ^_  G67R `<5^_667R

References

`55=5:5X5855555<5655=55 H5 :5 5 5 : ^_ /56 / R =67R `<@5>56555:5Y55^_:R 5L L=>9=67R `?5H/5`58`5X5X55>5-585X*5`/58G5-/5 V5 95 /<5 ^_ 5 R 85 X7R `=3553<5-5X<5X5`<G@5858<55<<5>^_XR /`G 6:7R `5 5 X5 65 `5 5 8@5 `5 5 5 `5 5 5 5 =<5 -5 55 >5 XG5 5 H<5 ` ^_ X G 68: ^-9R_ >Y5 /56 R=67R `5 `5 /5 65 85 ‹5 5 5 6G5 65 =5 85 5 = ^_  >5 7R L

Y555:565:5=5<5=65 5:5<5:58G585 5=5Y85:563555`5595-58@*565X5 : ^_  G :<58857R Y5 6 5 85 =5 X5 ^_  R 5-7R Y5 =5 G5 =5 `@5 H:5 -5 =5 5 =<5 85 />5 =@5/5=@5555/^_-9G @58<R

References

Y55@@55Y5`55-58@5:5855Y55X65@5 :^_6Y5-5X6 ^_G=6R Y565=5=55/555585-5‹5695`55Y5:5`5/5 95 Y5 5 5 5 5 Y5 5 Y@5 :>5 X5 565 =5 /6 ^_ - G ^9<_L9<_RG>6=7R Y55X5^_8-7-X67R Y55<555:5Y5`Y55:^_R @ G_5 5 /56R 67R Y5 65 5 55 X5 5 65 85 5 =:5 `G5 8 ^_ /5L/56 =>95 67R Y5585558555`5G5`55-5585-5Y5/@5 5 85 5 85 85 X5 = ^_ 9H: _:< 58857R Y5 5 85 85 Y5 5 `5 8/5 `5 / ^_ : G RR  6 7R Y5 5 5 -5 /@5 `5 /5 -- ^_  G7R Y5X55H55`55^_ 67R Y565=585565/<5855>5=5-585 =5/55><5655@@55@55X56`5=56^_5

References

G@5>55>9965 <7R Y5 :5 `5 -5 H<5 =5 5 Y5 -5 -55 <@5 5 @5 <55 65

Y5 /5 /5 ‹5 5 55 5 : ^_ >/ R 6 HYR X7R

YG5=5=5`:5:555‹X55=585-55>5 5-:5/5=85/5X55=R5=58<556@5 85 5 5 =5 :`5 `5 Y5 -5 -55 @5 5 5 5 -5 5 565=5=5565/55=5555855X585 <5 =5 ^_ 8 G <RG<67R

YG55559565/Y5@G5585-85=@5-5`5 55X5<55=55555655/5=55 -5/5 =5X5 Y:595 -55 855 >585 ^_ >=88-G?= 67T

Y55=`55556Y5H5><^_-@G =>9=67 R Y5 -5 =5 `Y5 Y@5 5 85 / ^_ -9R5 RRR X7R Y5 `5 =5 55 -5 -`5 85 95 -5 ^_ 6 5<7T

References

Y5 5 85 95 65 5 85 55 X5  ^_  <7R Y5 5 <@5 55 5 85 65 5 85 95 85 55 X5 Y5 X5  ^_ <G867R M

=585@5555Y5Y5-^_9< @R7R =5Y5@5:5`@5-:5YG5-^_5R >5 < G R67R =5 :5 65 `5 5 5 5 5 5 5 5 85 @5 -55 H5 >5 5 5 5855`5555=55=@=56565:5565Y^_  G 6 R  867R =5 5 /G@5 65 XX5 :5 5X5 =5 85 =5 Y:5 65 Y5 65 `5 /5 Y5 -5 /5 ^_  7 6  X7R =5>=5X5-^_ 6H7R =B5-55:55@5=5Y555X5-585=^_ : R67R =B5-5/_555Y5855Y555`5`<5855R/585YG5 `5X5Y5555<R5<5‹5/58=585>5585=5 G58>^_= 7R

References

=5555‹5@55555Y5/5=55/555Y55‹ ^_ 8 /56 8-9R 6 R 687R =5 =5 65 X55 55 Y/ ^_ X G7 55 5 6-7R =@R5 5 :5 > ^_ 6 67R =5=5`5-5_59^_67 H7T =5=5:5=55555<565-5=/5X55 5 85 `5 -5 _5 9 ^_ : 5 :<5 8857R =5=5`5-5_59^_67 7R =<5 55 :5 85 -5 =5 R/5 5 8G5 5 95 5 /5 85 5 >85 -5 = ^_ >6 L68:R R 7 @58:7R =5 X5 65 ` ^_ - R 5-7R =5 55 8<5 5 85 5 85 55 5 5 8R5 5 H5/5`55<5<5`G585/5855=^_ G<GG/56R>=H 7R

References

=5 8 55 Y5 :5 85 =5 H5 ><5 =`5 55 65 55 8G5 5= ^_ > @ @ =67R =X5 8 Y5 5 5 /5 `95 =5 5 8<5 5 `5 >5 /5 = ^_ 5=:R @ XY G G7R =`5 8 -5 G5 >5 55 `Y5 /@5 X5 5 5 G5 /5 `<5=595555X>5:565=@5-5^_ :G>=H7R =`5556Y5H5><^_5R GLR@RG RR:<58857R =`55556Y5Y55H5><^_8R6 <7R =`5 55 5 6Y5 H5 >< ^_  R 6_=67R =`5 5 ^_  G G 6=@6-7R ==585>G55585X5=5585‹555<5=55 /5 5 = ^_  :6 G R>=H7R =5 -5 `5 X5 5 Y5 5 = ^_ 59R@ 9@ RG-:`@<7R =55>G5555/55=^_:67 @57R

References

=5= =5_<585-@=5`5 5 :55-55<5X5 5=5>5-=5=5=585-^_6 R 9HH  6 7R =5==585-^_=GG >=G7R =5==585-^_=GG >=G7R =@5>55`565Y5>5<585:55`<5^_/56 =>9  >=H 7 R =*@5 /5 =5 5 X5 55 `*5 =5 :5 5 85 55 @5 55 :5 5 H5 :85 =@5  ^_ -_ Y8-9L-9  8 7 R =5 `5 8@5 =/ ^_  5 R 557R =5Y55-5=5>5=5`5-56565=5Y55 :5 5 `5 `55 65 55 /5 5 =`5 5 ^_ 5 @ GG@ @9>8Y7R =R-<5 >5 Y5 /5 @5 ` ^_  R G<7R =5 55 85 655 6<5 =55 : ‹<5 -5 5 >5 8 85 5 5 55 5 8-5 5 YX ^_  6 /565 /56 /56 G@/7R

References

=5=55<5Y65=5`565655X5555=5 =5X5Y^_=8-@6 7R =555=5Y5565^_><6- 8=-X7R =555X55Y59595/5/5658565:^_ <-<57R =5=5555@5``5/5/5>5-=5Y555X5Y ^_  G R < 5 G8X7R N

<55/5/55/5=5/5=5<5585/5<5 X5Y5555>5^_X55R65 56U@R7R <58555=Y^_:<5GG >5=6 7R <5 =5 XG5 5 9=R65 65 =5 <5 65 5 ^_ : 6  < 7 R <5=5H<5/5Y5<^_/56GRG7 -=H7R <5 8 5 5 5 5 5 X5 ` ^_ XR R @@R 7R

References

<5 `5 /5 65 @5 55 85 55 95 - ^_ 6 G=7R <5`:5H;<5Y:5=^_/7 G6HX 7T

<5 5 5 > ^_ >= 8- G R @ R ?67T

<58Y5`5Y^_ 7R O

H<558<5885Y<5-5565=5-G<5Y585>5`@55 X5=X5-<5=585/:5/58585-/5:@585555=5<5 5 5 >=R5 /5 5 :5 @5 5 ^_ 5 @R R6R=67R H@5855@555/GG5=Y565Y=55=^_: -7R H-5=5X5-^_6G >9>57R H5 /5 5 5 85 /5 G5 5 =5 5 @5 ^_ -6 67R H55-@555=59^_=7 7R P

:5 :5 55 Y5 `@5 /5 /5 <5 85 5 /5 X95 5 -5 85 85 5 Y ^_ 8R R

References

GG@867 R :55:5`^_>6 7R :@5-55=^_Y-<5G 7R :@5855555`5/5/58585Y5/5:@56X5`5^_ /5GH-7 R :5 X5 =<<5 =5 /5 65 85 X5 85 65 5 6 ^_ 6R R =67R :5 X5 65 <5 -5 65 X5 :5 :5 =5 95 65 5 95 H5Y58@*56556^_::5/56 =67R :5 =5 `5 /5 =`5 55 H5 ><5 5 > ^_ : @ /56 6 R  67R :5 =5 =5 5 5 > ^_ = /565 5 G  XR :5=55>^_- 6H:7R :5 Y5 85 > ^_  G /56 /G>6:7R :

References

:5`5X<595/5Y5/5Y56585X5=555^_ 6R :5 -5 85 =5 5 `5 <5 /5 R95 - ^_ 6Y =7R :G55<585`<5=5=`58-5=@5-5555=Y^_ : G @  =67R :G5 =5 <5 Y5 Y@5 5 =5 `Y5 G5 5 5 >5 5=5X5-555585/^_>9 :6 <7R :5=55/555=5585=5@55=555-555 RG5-5H5><^_/=>9 RG67R :565/5=Y5:5=55=5@55565Y=585X`5 <5 8Y ^_  :6  7R :565=5Y^_-67R R

-5 -5 -5 Y5 5 5 65 5 5 X5 -5 85 65 Y85 5 `5 =5-5Y5=55-558555Y55>5:565<5595Y5:5 55^_:7 87R

References

-5/5:5=5/@585=@555>5=5^_- Y-G596_/:Y87 -=5-5X5-5`R5=556^_87 655 77 -5 85 <@5 <5 85 5 5 Y5 85 - ^_ 5 G9`/- 667R -3<5 5 Y@5 `5 /5 ^_ 9 7 6H7R -585:5=55585=5`55X55X5-565=5 =@5 5 ^_ R <5R5-< 7R -5 95 -G5 >95 -RY5 55 `5 >=5 =: @6-7R -5 <5 6@5 =55 =5 : ^_ = /56 556=Y87R -RR5 -5 5 5 5 5 5 /5 >G5 55 5 =5 8@5 Y5 @5 5 85 55 <5 =5 5 = ^_  R :6 >=H 7R -55555-5`Y58=5/55>56558585 58585/^_-G RG@<7R

References

-5 <5 H85 55 /5 ^_  7 67R -5 : X5 =5 685 YG5 :5 /=@5 -5 8@5 5 5 =55 =5:5 858558555`6^_ : 6 G Y@-7R -3=5 X5 5 X5 5 5 > ^_ :R <5<5 -7R -@555=59^_RY7R -@5 5 X5 85 65 5 =5 <5 8Y5 =5 9 ^_ 9 R-7R -G5 95 5 95 95 5 @5 =5 :5 5 /@5 ^_  _ G RL: 6 6 R R 7R -5 8 5 5 =5 55 6 ^_ /  5 - 7T -G5 5 5 Y5 /5 :5 5 55 6 ^_ = G@G<-=67R S

8@5 <5 `5 85 `5 =5 5 :5 =5 9 ^_  R 6 GR5 6R@ >=H7R

References

85>58@555X5/5Y5<@5-555>Y5Y5/5:5/:5 X@5858558G55H5X58G5-5@5:55= ^_6Y76Y5 7R 85Y-55>^_87G H7R 8G5 55 ==5 85 85 X5 5 55 5 X55 5 = ^_:6@ 5557R 85=59<=585/555`@5-5`5X5Y5 >5 5 =5 =5 -/ ^_ 6 G G 96H @ =67R 8585`<5-5X5^_-G G65<7R 85 -5 85 5Y5 8<5 /5 5 /5 85 Y=5 85 / ^_ X 6R 7R 85 > 5 85 =65 55 5 85 / ^_ -_ :87R 8X565658^_-7R /56/5677 8565:55=<<5659585=55-5=55-55 `55858565:5:Y^_9 G=``LX9L:`L5` =67R 85 Y5 <_<5 <5 5 ^_ 6 5 >6=^=_7R

References

85 55 :5 55 85 =8 ^_  @ RR7@ 7R 8555Y5`5`56555/5>=5-555X5=55Y55 /5 65 `5 85 Y= ^_ R G 667R 8G<5=5X5=^_9R 5-7R 8R85 =5 Y5 `5 =/<R5 Y5 Y5 5 =/<R5 =X5 65 ` ^_ R _ G R7R 85 - :5 X5 -5 X5 : ^_ / 6 6<G=7< G@8<-7R 855Y585X5`=^_>@G 57R

85559585>55‹^_>8-55 G?6 7T

85`585:555<@585H`5=^_: 57R 85 X ^_ Y G 6685X=75 8<5-58@5-5`<555=565/55555Y5655555 5565>55585655>55-5Y5>=55-5 85X5<5=5`5=585-5=-5>55Y^_8 /56  87R

References

85 65 5 85 5 65 `5 ^_ / ^/56_ G6^==:_ 667R 85-5858^_:G@/ <-667R 8R85 85 5 5 <5 8Y5 `5 -:5 65 9 ^_ - G=<7 R 85 Y5 5 5 `5 H5 95 85 Y 5 < ^_ /56 GRG5H7R 85 5 `5 85 =5 Y<5 5 ^_ Y R G 6L>: R :Y8H7 855556^_< 7R 85665Y5855=5Y5Y^_8/56/6  6=`L:` ::5 5 : 6 : 76R 855?@5/=5Y5X>^_6< ^=>9_ 6 7 =>96 R =7R T

55@5:5@585`5=85H<5`^_ _ G -9R  7R 5 5 5 /5 5 8 ^_ / 68T

References

55X585X55=Y5`>55Y5-5565>^_ 5R  6 7 R 5-55>95H<@@5>565655Y5:5:=5Y55@5Y55 H5 ><5 X5 =5 5 55 /@5  ^_  G <=7R 55/56558G58Y^_5 -87R 5 =5 65 Y5 /5 5 `5 /X5 X5 /5 /5 5 `5 65 ^_ >GR R :<58857R 5 Y5 =X=5 65 =5 5 5 <5 XX5 <5 5R/5 X5 :5955:5@5=5<R5^_-8- GGGL>=H7R 58565`/5Y5/5=5<5555:59^_ /7_5< 58T 5H5555^_/=X5<5RG 87R 5=585>^_@7 <@-X7R 5>656GG5858GG5:5=6585:55YG555 5 5 <5 X5 `5 5 /5 85 85 95 5X5 =@5 = ^_ @ <5RG R R7R 5 85 `5 85 `@5 85 @5 H5 =5 -5 <G5 55 =5 ^_ /56 85 G ::5-67R

References

5 5 Y5 5 5 <-5 5  ^_  /565 R<5-7R 5 85 >RH5 55 65 =>5 @5 6 ^_ =G G 7<-<7R 555=^_<5U^_ 6 R7R 5555555565X5-^_Y>9R5 R/=X5R `aX7R G55=5>=5:585XV559585Y5>5X5-^_> @G55R 7R U

5=56@5555585`G5-5*5Y5X<585^_- 8-G<5-^5_7R G<@5:55‹59@555>955=^_6G@ G@ :6L85:67R V

Y5 5 =5 <5 `=5 85 : ^_ `9 /56 >=H7R

 X5 95 X5 =5 X5 65 `5 5 :5 ::5 5 55=@5 -55595855Y5H^_><5 R?<=7T

References

H5 65 `5 >5 X5 ^_  R 6HX7R 55556595X565>655=5-<556@5 :5:56595-5X5:58@*565=58^_6 @<R5/5656 6G:<58857R

<5/558`5<5>>558595-55:5`5X5Y5G5-5 ^_>8-^8-_<5R?67R

5-5=5`5H55G556555=5>5=555:=<565 85 =5 -5 55 `5 X5 H5 >< ^_ / @67R 5-5/565=5>5-G56-5:5:5H5><5=`5 5 ^_ : @ 6 R 6=67R 555`G58^_@ 66R 67R 5 >5 _5 95 ` / X ^ _ 6 7 X7T 5>5_59595595-5=55<5^_= =><7R 5>5`585Y55Y5-^_-R7 -7R W

55XX5558Y5<555>5655565 ^_ / /56 R 958>7R

References

55/55<-5<5=5:5<565=5/5//555/5 5 5  ^_ /565 /55 =67R 55/5`@5=5555<-5<5=5:55<R5:5 5^_-GGRR= 67R 55=5<5<585/5/5=5==585-55-5 95 /5 5 5 =5 = ^_ 5 R G67R 5 65 95 =5 55 -/5 8R<5 Y5 -5 55 5G5 65 Y5 =:5 `<G5 85 `5 85 /5 5 9_5 855 Y<5 X<5 `5 :5 :5 -X ^_  G 67R 5/5Y56/5‹5655:59555GG585555=5/6^_ -9R557 R 5/5<5555Y585Y5855/5G5=5X5 `^_8 @7R 5556565`585>5@555:555`^_XR /5/56 67R 5`55=5H<55=55H@5=55G555<5585 / ^_ 5 /56 X55R =>Y  H7R 55-V@555X@55@55>G5=:5:5=55=5@5 65 -V@5 6 ^_ 5 6 7  Y H 7 R

References

5>55G5585=^_R7 87R 55H555/<555G55Y565`5=/5-5:5Y55 5-5^_=GR @67R @5:5`5=585==55G565<@5H58<55=5 `= ^_  = _ <7R 5 85 G5 ^_  G G8-7T X

5 /5 5 =5 95 -5 X5 ^_ 8 67R Y

5/55>X5=5Y56S5659555=5/X565 5 X5 5 5 >5 -=5 8=5 `5 565 ^_ <6- 667R 5585>^_/5/5675 H7R 5585>^_  -L/ 7G <-=67R 5585>^_  -L/ 7G <-=67R

References

5  /5 <G5 Y ^_ / G 6:><-5 :-6:7-R 5  Y5 5 =5 85 > ^_ -  G =67R 5=5X5^_>6H7 R 55<585R:@565X5`^_> G@<7R 5 5 :5 Y5 -@5 -5 85 > ^_ / ^/56_ 56^-=_67R 5Y5X556=5565> ^_8 < G G87R Z

5 85 =5 -5 5 :5 G5 5 5 5 85 = ^_ >9RG@ 7R 5855G555:58=G55‹5/585=^_8R R :<58857R @5 =5 8@5 -5 95 5 5 85 65 ^_ / /56 G GRX7R 5 6 Y5 =`5 55 65 85 55 6Y5 /5 55 H5 >< ^_ 6 R67 R

References

5‹55/55<5X5-/^_5>5 G 6: : < 5 8  8 5 7 R 56Y5=`555H5><^_>5 7 ?=67T

5555555858R555`55H5<5:5 5 5 5 5 8-5 5 :5 Y5 85 85 > ^_ /56 = 6 =G55Y6=67T 5585:5G5/^_6RR6>5_ 67R 5=^_9>8Y 7R @<5X555:5=58<@55-5`5>5X5XG5-5 85<56585@55856^_< 6G=67R

ANNEX1 LISTOFFIGURESANDABBREVIATIONS

Listoffigures

Listoffigures LISTOFFIGURES

Introduction

97=

976

97:

97

97-G

978-R

97/56=>9==:6

976/56<

97-G<G

Results

9-7=R@

9-7/566R R

9-7/56R R

9-7/56RR

Listoffigures 9-7-R_:6-66Hdac7,Pax5, Rag1Csf1r

9-7R/56

9-7/56R6

9-7X66 =9

9-7/56R6

9-7-R_:6-

9-7/56R6

9-7/56@@R

9-7/56G=>96

9-7/56=>96R

9-7/56G=>96 @R

9-7/56R6

Listoffigures 9-7/56R6=6

9-7/56R6

9-7/56R6=

9-7/56R6

9-7/56=>96 =R

9-7/566G

9-78@@ /56

9-7/56@@

9-7R

9-7G G/566:L@

9-7/56

9-7/56@@

Listoffigures 9-7G /56@@

9-7=<5 GG/566:L@

9-7/56

9-7/56R

9-76@ G/56

9-7G_:6-6 5G/56

9-7/56RR R

97/56G=>96 R

9-7:R@/56RYR-6 56R

9-7:R@/566=R

G-766 5G/56

Discussion

97=/56RR

Listoffigures 97=/56RR R

Abbreviations

Abbreviations

ABBREVIATIONS

5YY 5G@ 6- R = : <5 66G 6: 6 6Y: 6 6=: 6 6-6 66 68- 6RG <5 6G 9H 9 X/ X X=: XL XH X /5 / / / /56 / / /G /86 / =G `H `@R Y=:: = Y:8 Y =G =G =>9 =

Abbreviations

=>: =@L =9 = =9 = = ^R^5R<RR_R5R<G = =< <>8 <6 <` <@

Abbreviations

:5     

ANNEX2 PUBLICATIONS

LISTOFPUBLICATIONS

:/56-=X-_ :R6=; R5 5 RomanGonzalez, L.,* 6<5 H5 -5 /5 5 5==`5 5 Y/5 5 55 55 Y5 X5 X5 YS

:> 5 5 ; R5RomanGonzalezL56<H5=5:= CompFunctGenomics.67

:/56G @; R5 5 6<5 H5 85 5 5 5==`5 X<5 55 Roman Gonzalez,L5555Y5=5:5= =G

:/5675; 5G5/56 G in vivo @@ /56G G/56G@ G8=

HDAC7 Is a Repressor of Myeloid Genes Whose Downregulation Is Required for Transdifferentiation of Pre-B Cells into Macrophages

Bruna Barneda-Zahonero1., Lidia Roma´n-Gonza´lez1., Olga Collazo1, Haleh Rafati2, Abul B. M. M. K. Islam3¤, Lars H. Bussmann4, Alessandro di Tullio4, Luisa De Andres4, Thomas Graf4,5, Nu´ ria Lo´ pez-Bigas3,5, Tokameh Mahmoudi2, Maribel Parra1¤* 1 Cancer Epigenetics and Biology Program (PEBC), Bellvitge Biomedical Research Institute (IDIBELL), Barcelona, Spain, 2 Erasmus University Medical Center, Department of Biochemistry, Rotterdam, The Netherlands, 3 Research Unit on Biomedical Informatics, Department of Experimental and Health Sciences, Universitat Pompeu Fabra, Barcelona, Spain, 4 Center for Genomic Regulation, Barcelona, Spain, 5 Institucio´ Catalana de Recerca i Estudis Avanc¸ats (ICREA), Barcelona, Spain

Abstract B lymphopoiesis is the result of several cell-commitment, lineage-choice, and differentiation processes. Every differentiation step is characterized by the activation of a new, lineage-specific, genetic program and the extinction of the previous one. To date, the central role of specific transcription factors in positively regulating these distinct differentiation processes to acquire a B cell–specific genetic program is well established. However, the existence of specific transcriptional repressors responsible for the silencing of lineage inappropriate genes remains elusive. Here we addressed the molecular mechanism behind repression of non-lymphoid genes in B cells. We report that the histone deacetylase HDAC7 was highly expressed in pre-B cells but dramatically down-regulated during cellular lineage conversion to macrophages. Microarray analysis demonstrated that HDAC7 re-expression interfered with the acquisition of the gene transcriptional program characteristic of macrophages during cell transdifferentiation; the presence of HDAC7 blocked the induction of key genes for macrophage function, such as immune, inflammatory, and defense response, cellular response to infections, positive regulation of cytokines production, and phagocytosis. Moreover, re-introduction of HDAC7 suppressed crucial functions of macrophages, such as the ability to phagocytose bacteria and to respond to endotoxin by expressing major pro-inflammatory cytokines. To gain insight into the molecular mechanisms mediating HDAC7 repression in pre-B cells, we undertook co- immunoprecipitation and chromatin immunoprecipitation experimental approaches. We found that HDAC7 specifically interacted with the transcription factor MEF2C in pre-B cells and was recruited to MEF2 binding sites located at the promoters of genes critical for macrophage function. Thus, in B cells HDAC7 is a transcriptional repressor of undesirable genes. Our findings uncover a novel role for HDAC7 in maintaining the identity of a particular cell type by silencing lineage- inappropriate genes.

Citation: Barneda-Zahonero B, Roma´n-Gonza´lez L, Collazo O, Rafati H, Islam ABMMK, et al. (2013) HDAC7 Is a Repressor of Myeloid Genes Whose Downregulation Is Required for Transdifferentiation of Pre-B Cells into Macrophages. PLoS Genet 9(5): e1003503. doi:10.1371/journal.pgen.1003503 Editor: H. Leighton Grimes, Cincinnati Children’s Hospital Medical Center, United States of America Received July 5, 2012; Accepted March 25, 2013; Published May 16, 2013 Copyright: ß 2013 Barneda-Zahonero et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Funding: This work was supported by a grant from the Spanish Ministry of Science and Innovation (MICIIN) (BFU2008-00410) (to MP), a grant from the Spanish Ministry of Economy and Competitiveness (MINECO) (SAF2011-28290) (to MP), and a Marie Curie IRG grant from the European Commission (FP7-MIRG-CT-2007- 208119) (to MP). MP was supported by a Ramon y Cajal contract from the Spanish Ministry of Science and Innovation (MICIIN). LR-G was supported by an IDIBELL PhD fellowship. ABMMKI was supported by fellowship from AGAUR, Catalonian Government, Spain. TM was supported by an Erasmus MC Fellowship. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. Competing Interests: The authors have declared that no competing interests exist. * E-mail: [email protected] ¤ Current address: Department of Genetic Engineering and Biotechnology, University of Dhaka, Dhaka, Bangladesh . These authors contributed equally to this work.

Introduction lymphoid progenitors (CLPs), which have the potential to differentiate into B and T lymphocytes, as well as natural killer The generation of B cells is the result of several cellular (NK) cells [3]. The transcription factors IKAROS, PU.1 and transitions that take place in a stepwise manner and comprise cell MEF2C are critical for the cellular commitment of LMPPs to the lineage choices, cell commitment and differentiation. Every lymphoid lineage [3–5]. Later, the transcription factors E2A, EBF differentiation step leads to the activation of specific genes and FOXO-1 are required for the early specification of CLPs into characteristic of the new cellular stage. This is achieved by the pro-B cells, whereas PAX5 is required to maintain B cell identity action of well defined networks of transcription factors specific to along differentiation into mature B cells [6–11]. However, there is each particular cellular state [1,2]. In the bone marrow, an increasing body of evidence indicating that the repression of lymphocyte development begins at the lymphoid-primed multip- lineage inappropriate genes is a pivotal mechanism to properly otent progenitor (LMPPs) stage. LMPPs become common acquire a particular cellular state during B lymphopoiesis. For

PLOS Genetics | www.plosgenetics.org 1 May 2013 | Volume 9 | Issue 5 | e1003503

HDAC7 Is a Key Repressor in Pre-B Cells

b Author Summary retroviral vector for stable expression of a -estradiol-inducible form of C/EBPa (C10 and C11 cells). After addition of b-estradiol, Through the hematopoietic system, all the distinct mature C10 and C11 cells are converted into functional macrophage-like blood cell types are generated, thereby constituting one of cells at 100% efficiency within 48–72 hours. The conversion of the best-studied paradigms for cell lineage commitment pre-B cells into macrophages is direct and does not involve overt and differentiation in biology. B lymphocytes are gener- retro-differentiation through hematopoietic stem and progenitor ated through several cell-commitment, lineage-choice, and cells [18]. Unexpectedly, this cellular transdifferentiation process differentiation processes. To date, the central role of appears to occur in the absence of significant changes in the DNA lineage-specific transcription factors in positively regulat- methylation status of key lymphoid or myeloid genes, but involves ing these distinct developmental steps is well established. changes in histone modification on both types of genes [19]. Since However, in the absence of proper transcriptional repres- HAFTL cells are a fetal liver cell line immortalized by Ha-ras sion, an ‘‘adolescent cell’’ will never be able to reach its transformation we, in parallel, also investigated the involvement of ‘‘adulthood identity,’’ having a potential impact in the development of hematological malignancies. In this article, class IIa HDACs in normal primary B cell precursors. Here, we we examined the molecular mechanism responsible for report that during the conversion of pre-B cells into macrophages, the gene silencing of lineage undesirable genes in B cell HDAC7 expression is down-regulated. In pre-B cells HDAC7 precursors and uncovered the role played in this process specifically interacts with the transcription factor MEF2C and is by the histone deacetylase HDAC7. We show that HDAC7 recruited to putative MEF2 sites located at the promoters of key is expressed in B cell precursors where it interacts with the macrophage genes. Forced re-expression of HDAC7 interferes transcription factor MEF2C and is recruited to the with the establishment of the gene transcriptional program and promoters of non-B cell genes. While HDAC7 is down- functional characteristics of macrophages. Importantly, HDAC7 regulated during the lineage conversion of pre-B cells into depletion in pre-B cells results in the de-repression of macrophage macrophages, re-expression of HDAC7 interferes with both genes. the acquisition of the myeloid gene transcriptional program and macrophage-specific cell functions. We Results therefore have identified a novel lineage-specific tran- scriptional repressor in the hematopoietic system. HDAC7 is down-regulated during the transdifferentiation of pre-B cells into macrophages example, PAX5 not only induces the expression of a B-cell specific In order to study the potential role of class IIa HDACs in B genetic program, it also suppresses inappropriate genes of lymphocyte biology we first analyzed their expression levels during alternative lineages, thereby ensuring its role in maintaining B the cellular transdifferentiation of pre-B cells into macrophages. cell identity and differentiation [12–14]. Recently, it has been RT-qPCR and Western blotting experimental approaches showed reported that the transcription factor MEF2C, by activating that HDAC7 expression was dramatically down-regulated during lymphoid specific genes and repressing myeloid genes, is involved the conversion of C10 cells into macrophages at 72 hours after b- in the cellular choice towards the lymphoid lineage [5]. These estradiol treatment (Figure 1A and 1B). Notably, no changes in studies suggest that B cell transcription factors must also recruit Hdac4, Hdac5 and Hdac9 expression levels were observed during transcriptional co-repressors to silence undesirable genes. To date, the cellular reprogramming process. As previously described [17], very little is known on the role of transcriptional repressors during the B cell genes Rag1 and Pax5 became down-regulated, whereas B lymphopoiesis. the expression of the myeloid genes C/Ebpb and Csf1r were up- Histone deacetylases (HDACs) have emerged as crucial regulated during cellular reprogramming (Figure 1A). Consistent transcriptional co-repressors in highly diverse physiological with our findings, HDAC7 is not present in RAW264.7 systems. To date, 18 human HDACs have been identified and macrophages at neither the RNA nor protein levels (Figure 1A grouped into four classes. Class I HDACs (HDAC1, 2, 3, and 8), and 1B). As expected, the expression of the B cell transcription class II HDACs (HDAC4, 5, 6, 7, 9, and 10), class III HDACs, factors IKAROS, E2A, EBF and PAX5 were down-regulated also called sirtuins, (SIRT1, 2, 3, 4, 5, 6, and 7) and class IV during the cellular conversion. In contrast, protein levels of the HDAC (HDAC11). Class II HDACs are further subdivided into transcription factors MEF2C and RUNX1 did not significantly class IIa (HDAC4, 5, 7, 9) and class IIb (HDAC6 and 10) [15,16]. change during transdifferentiation (Figure 1B). To further confirm Unlike other HDACs, Class IIa HDACs have three unique our findings we analyzed the kinetics of HDAC7 expression during features. First, they are expressed in a tissue-specific manner and the cellular transdifferentiation in our previously reported are involved in development and differentiation processes. They microarray experiments [17]. Strikingly, similar to the B cell exert their transcriptional repressive function in skeletal, cardiac, specific transcription factor PAX5, HDAC7 expression was down- and smooth muscle, the bone, the immune system, the vascular regulated during cellular reprogramming (Figure S1). Since system, and the brain among others. Second, they are signal- HAFTL cells are a fetal-derived pre-B cell line transformed by dependent co-repressors which become phosphorylated at con- Ha-ras we wondered whether our findings could be extended to served serine residues in the regulatory N-terminal domain leading normal primary cells. We analyzed HDAC7 expression in both to their nuclear export. Third, they contain a regulatory N- primary B cell precursors and primary macrophages in a recently terminal domain that mediates their interactions with tissue- reported microarray analysis [18]. Importantly, we found that, specific transcription factors such as members of the MEF2 family similar to PAX5, HDAC7 was highly expressed in B cell [15,16]. This last feature prompted us to ask whether members of precursors compared to primary macrophages (Figure 1C and the class IIa HDACs subfamily could be lineage-specific Figure S1). In agreement with our results with the C10 cell line, transcriptional repressors crucial to maintain B cell identity and C/EBPa-mediated conversion of primary pre-B cells into macro- biology. phages also resulted in the down-regulation of HDAC7 (Figure 1C To address this question we have used a cellular transdiffer- and Figure S1). Our data demonstrate that among the different entiation system that we reported recently [17]. This system class IIa HDACs, HDAC7 shows a lymphoid-specific expression consists of a pre-B cell line (HAFTL cells) transduced with a pattern and suggest that by repressing lineage inappropriate genes

PLOS Genetics | www.plosgenetics.org 2 May 2013 | Volume 9 | Issue 5 | e1003503

HDAC7 Is a Key Repressor in Pre-B Cells

Figure 1. HDAC7 is down-regulated during the transifferentiation of pre-B cells into macrophages. A. RT-qPCR experiments for gene expression changes of class IIa HDACs, B cell and macrophages genes (uninduced cells (2) and b-estradiol induced C10 cells (+) for 72 hours). HAFTL

PLOS Genetics | www.plosgenetics.org 3 May 2013 | Volume 9 | Issue 5 | e1003503

HDAC7 Is a Key Repressor in Pre-B Cells

pre-B cells and RAW264.7 cells were used as control. Data are represented as the mean +/2 standard error of the mean (SEM) of three independent experiments. B. Western blot of Class IIa HDACs and B cell transcription factors in the same experimental conditions shown in A. C. RT-qPCR experiments for gene expression changes of Hdac7, Pax5, Rag1 and Csf1r genes (uninduced (0) and b-estradiol induced primary pre-B cells transduced with a retroviral vector for C/EBPa expression for the indicated times. Primary macrophages were used as control. doi:10.1371/journal.pgen.1003503.g001 in pre-B cells, such as genes characteristic of macrophages, regulation is impaired by HDAC7 we found several that are HDAC7 might be crucial in maintaining B cell functions and involved in phagocytosis (such as Fcgr1, Fcgr2b and Fcgr3), genes identity. related to the immune response (such as the chemokine genes Cxcl10, Cxcl11 and Ccl2), Toll-like receptors (Tlr3, Tlr4, Tlr7, Tlr8 HDAC7 re-expression interferes with the gene and Tlr9), interleukins (Il18 and Il15) and TNF pathway-related transcriptional program of the reprogrammed genes (Tnfsf10, Tnfsf11 and Tnfsf13b) (Figure 3). A full list of the macrophages genes found in each category is presented in the interactive website http://bg.upf.edu/C10-HDAC7/ (for statistics see Dataset S1). Our observed dramatic down-regulation of HDAC7 upon To validate our microarray analysis, we selected several represen- transdifferentiation of pre-B cells into the macrophage lineage was tative genes and tested their mRNA levels by RT-qPCR. We consistent with its potential role as a transcriptional repressor of observed that the presence of HDAC7 significantly interfered with macrophage-specific genes in pre-B cells. To test whether the the increase in the mRNA levels of Cxcl10, Fcgr1, Ifi35, Ifi47, Tlr3, presence of HDAC7 could interfere with the acquisition of a Tlr9, Il18 and C3 genes after b-estradiol treatment (Figure 4). macrophage-specific gene program during transdifferentiation of Importantly, HDAC7 re-introduction did not interfere with the pre-B cells into macrophages, we performed a gain of function down-regulation of key genes for B cell differentiation and biology experimental approach followed by genome-wide microarray such as the transcription factor PAX5, reinforcing the notion that analysis. We transduced C10 cells with a retroviral vector carrying HDAC7 is a repressor of macrophage genes in B cell precursors HDAC7-Flag (C10-HDAC7 cells). As a control, C10 cells were (Figure S5). During the trasdifferentiation of C10 cells into transduced with an empty retroviral vector (C10-MSCV cells). As b macrophages the cells stop dividing and many genes related to the expected, -estradiol treatment of C10-MSCV cells resulted in the cell cycle, such as genes involved in mitosis, become down- total down-regulation of endogenous HDAC7 protein levels, while regulated [17]. We have observed that the presence of HDAC7 C10-HDAC7 cells expressed the exogenous HDAC7 protein at also interferes with the down-regulation of these types of genes, similar levels as in untreated C10-MSCV cells even at 72 hours b corroborating that HDAC7 blocks, at least in part, the cellular after -estradiol treatment (Figure S2). We then examined the transdifferentiation process (Figure S4, Dataset S2, and interactive genome-wide effects of HDAC7 re-expression on the gene website http://bg.upf.edu/C10-HDAC7/). Our data clearly transcription program of C10 cells induced to transdifferentiate demonstrate that HDAC7 represses the expression of central to macrophages. Microarray experiments were conducted in both genes for macrophage function, and strongly suggest that HDAC7 C10-MSCV and C10-HDAC7 cells un-induced or induced to acts as a specific transcriptional repressor of lineage inappropriate b transdifferentiate for 48 and 72 hours. The addition of -estradiol genes in B cell precursors. to C10-MSCV cells resulted in the up- and down-regulation of 1609 and 1798 genes at 48 hours and of 1531 and 1567 genes at HDAC7 knock down leads to the de-repression of 72 hours after treatment, in agreement with our previous report [17]. Importantly, the exogenous expression of HDAC7 in C10- macrophage genes HDAC7 cells treated with b-estradiol totally or partially abrogated To determine the physiological function of HDAC7 in B cell the up-regulation of 988 and 866 genes, after 48 and 72 hours precursors and test whether it is involved in the repression of respectively. Earlier analyses showed that the vast majority of the macrophage genes, we performed a loss of function experimental up-regulated genes correspond to key genes for macrophage approach. We knocked down HDAC7 by siRNA in both HAFTL function, whereas the down-regulated genes are associated with cells and primary pre-B cells (Figure 5). Strikingly, we observed cell cycle processes and with important functions for B cell that the reduction in HDAC7 mRNA levels resulted in the de- development and biology [17]. To determine the type of genes repression of key macrophage genes such as Itgam (Mac-1), Fcgr1 whose up-regulation was affected by the presence of HDAC7 we and Ccl3 (Figure 5). These data demonstrate that HDAC7 is performed a gene set enrichment analysis based on the gene involved in the repression of lineage inappropriate genes in B cell ontology (GO) categories corresponding to Biological Processes, precursors. Cellular Components and Molecular Functions, as well as on KEGG pathways. To do so we took advantage of Gitools, a HDAC7 binds to the transcription factor MEF2C in pre-B recently developed bioinformatics tool [20]. Gitools allows cells and is recruited to the promoters of target genes accessing many available biological databases, performing analysis critical for macrophage function and visualizing data using interactive heat-maps. Strikingly, the At the mechanistic level, class IIa HDACs exert their actions as Biological Process enrichment revealed that the set of up-regulated transcriptional repressors by interacting with specific transcription genes affected by HDAC7 belong to GO categories representing factors recruited to the promoters of genes required for key macrophage related functions, such as immune, inflammatory development and cell differentiation [15,16]. To address whether and defense response, cellular response to infections, positive HDAC7 specifically interacts with particular sequence-specific regulation of cytokines production and phagocytosis (Figure 2A). transcription factors in pre-B cells, we undertook a candidate Importantly, we found that the categories enriched in the KEGG approach and performed co-immunoprecipitation experiments to pathways analysis correspond to similar biological processes test for potential interaction between endogenous HDAC7 and the (Figure S3). Gene ontology (GO) analysis corresponding to B cell transcription factors IKAROS, E2A, PAX5, and MEF2C in Cellular Components and Molecular Functions reinforce the the C10 parental line (HAFTL cells). HDAC7 specifically above results (Figure 2B and 2C). Among the genes whose up- associated with MEF2C, but not with the other B cell transcription

PLOS Genetics | www.plosgenetics.org 4 May 2013 | Volume 9 | Issue 5 | e1003503

HDAC7 Is a Key Repressor in Pre-B Cells

Figure 2. HDAC7 re-expression interferes with the gene transcriptional program of the converted macrophages. Heat-maps showing significantly (corrected p-value ,0.05) enriched GO Biological Process (A), Cellular Components (B) and Molecular Functions (C) categories among the up-regulated genes affected by the re-expression of HDAC7 during transdifferentiation of pre-B cells into macrophages. Colours toward red indicate high statistic significance, yellow indicates low statistic significance, and gray indicates no statistical significance. The list of genes for the enrichment analysis in each column is as follows: C10-MSCV-48h includes genes up-regulated after b-estradiol treatment for 48 hours. C10-HDAC7-48h includes genes up-regulated after b-estradiol treatment for 48 hours which are down-regulated in the presence of HDAC7 (HDAC7 re-expression). C10-MSCV- 72h includes genes up-regulated after b-estradiol treatment for 72 hours. C10-HDAC7-72h includes genes up-regulated after b-estradiol treatment for 72 hours which are down-regulated after re-expression of HDAC7. doi:10.1371/journal.pgen.1003503.g002 factors tested (Figure 6A). Recently, Camargo and colleagues have (Figure 6B). Strikingly, the phagocytosis-related genes Fcgr1, Fcgr2b reported that MEF2C is crucial in the cellular choice towards the and Fcgr3 were found to be targets of both HDAC7 and MEF2C. lymphoid versus the myeloid lineage in LMPPs [5]. Microarray These data indicate that MEF2C represses genes characteristic of analysis of control and MEF2C-deficient LMPPs revealed that the macrophages in hematopoietic progenitors and suggest that absence of MEF2C resulted in the up-regulation of genes enriched HDAC7 may also silence lineage inappropriate genes at earlier in GO categories related to the immune system, such as genes stages of lymphocyte development. involved in the inflammatory and defense response of the cells [5]. To test whether in pre-B cells HDAC7 is recruited to the Comparison of these data with our set of genes whose up- promoters of macrophage-specific genes whose up-regulation is regulation was affected by HDAC7 re-expression during cellular impaired in the presence of exogenously expressed HDAC7, we transdifferentiation showed a significant overlap by Chi-square test performed chromatin immunoprecipitation (ChIP) assays. Nota- 2 of 46 (p,10 6) and 30 (p = 0.00044) genes, respectively bly, using the TFconsite bioinformatic tool we found that

PLOS Genetics | www.plosgenetics.org 5 May 2013 | Volume 9 | Issue 5 | e1003503

HDAC7 Is a Key Repressor in Pre-B Cells

Figure 3. HDAC7 re-expression interferes with the gene transcriptional program of the converted macrophages. Heat-maps showing observed differentially expressed genes for selected enriched GO categories. Blue colour cell indicates positive events while gray colour indicates that the gene was not observed differentially expressed in that experimental condition. doi:10.1371/journal.pgen.1003503.g003 promoters of Fcgr1, Cxcl10 and Itgam contain putative MEF2 S6A). Chromatin prepared from C10 cells un-induced or induced binding sites. We designed primers spanning the upstream with b-estradiol for 72 hours was subjected to ChIP with regulatory regions, gene body, and downstream regulatory regions antibodies specific for MEF2C and HDAC7. qPCR analysis of of mouse Fcgr1, Cxcl10 and Itgam (Figure 6C and 6F and Figure the immunoprecipitated material with primers specific for the

PLOS Genetics | www.plosgenetics.org 6 May 2013 | Volume 9 | Issue 5 | e1003503

HDAC7 Is a Key Repressor in Pre-B Cells

Figure 4. RT–qPCR for microarray validation of selected genes. RT-qPCR experiments for gene expression changes for 8 up-regulated genes, Cxcl10, Fcgr1, Ifi35, Ifi47, Tlr3, Tlr9, Il18 and C3 in the absence or in the presence of HDAC7. Data are represented as the mean +/2 standard error of the mean (SEM) of three independent experiments. The two-way ANOVA test was used to calculate significant levels between the indicated groups. *P,0.001. doi:10.1371/journal.pgen.1003503.g004

PLOS Genetics | www.plosgenetics.org 7 May 2013 | Volume 9 | Issue 5 | e1003503

HDAC7 Is a Key Repressor in Pre-B Cells

PLOS Genetics | www.plosgenetics.org 8 May 2013 | Volume 9 | Issue 5 | e1003503

HDAC7 Is a Key Repressor in Pre-B Cells

Figure 5. HDAC7 knock down leads to the de-repression of macrophage genes. HAFTL cells (A) and primary B cell precursors (B) were transfected with control siRNA or siRNAs targeting HDAC7. 72 hours after transfection Hdac7, Itgam, Fcgr1 and Ccl3 mRNA levels were determined by RT-qPCR experiments. Data are represented as the mean +/2 standard error of the mean (SEM) of three independent experiments. The two-way ANOVA test was used to calculate significant levels between the indicated groups. *P,0.001. doi:10.1371/journal.pgen.1003503.g005

Fcgr1, Cxcl10 and Itgam loci indicated that both MEF2C and shown that C/EBPa-mediated reprogramming of pre-B cells into HDAC7 were specifically and significantly enriched at the macrophages occurs in the absence of significant changes in the identified putative MEF2 binding sites in un-induced C10 cells DNA methylation of crucial genes suggesting that the repro- (Figure 6D, 6E, 6G and 6H and Figure S6B). Importantly, while grammed macrophages retained an epigenetic memory charac- significant MEF2C enrichment around its putative binding sites teristic of the cell of origin [19]. Therefore, we speculate that the was found in both un-induced and induced conditions, HDAC7 chromatin structure present in differentiated macrophages is not enrichment at the target genes was lost upon b-estradiol induction permissive for the recruitment of HDAC7 to its target genes and (Figure 6D, 6E, 6G and 6H and Figure S6B). Taken together these that such recruitment is only possible in a chromatin environment data indicate that via interaction with MEF2C, HDAC7 is characteristic of B cells, in line the notion that HDAC7 is a recruited to the promoters of myeloid genes in B cell precursors, lymphoid-specific transcriptional repressor. Taken together, these resulting in their transcriptional silencing. results demonstrate that HDAC7 significantly interferes with key functional characteristics of the transdifferentiated macrophages. HDAC7 re-expression interferes with the functional properties of the reprogrammed macrophages Interaction of HDAC7 with MEF2C and its catalytic We previously reported that during reprogramming of pre-B activity are essential for repression of Mac-1 cells into macrophages there is an increase in the expression of Class IIa HDACs interact with MEF2 proteins via a conserved Mac-1 and a down-regulation of CD19 levels, two cell surface motif of 17 amino acids located in the amino-terminal region of markers characteristic for macrophages pre-B cells, respectively. the proteins. To definitively prove that HDAC7 represses The reprogrammed macrophages show high phagocytic activity macrophage genes through its interaction with MEF2C, we and respond to LPS treatment with cytokine production [17]. To generated retroviral vectors carrying mutants of HDAC7 with test whether the presence of HDAC7 could interfere with the either a deletion of the entire 17 amino acids stretch (HDAC7- functional characteristics of the reprogrammed macrophages we DMEF) or with substitutions of crucial lysine residues (HDAC7- undertook three different experimental approaches. First, we K86AK88A). We tested the HDAC7 mutants for their ability to followed the expression kinetics of CD19 and Mac-1 (CD11b), two repress Mac-1 during the transdifferentiation of pre-B cells into cell surface markers characteristic for pre-B cells and macrophag- macrophages. Expression of wild-type HDAC7 resulted in a es, respectively by flow cytometry. In both C10-MSCV and C10- significant decrease in Mac-1 positive cells and mRNA levels HDAC7 cells, 100% of the population became CD19 negative (Figure 8) whereas expression of the HDAC7 mutants defective for and Mac-1 positive 72 hours after addition of b-estradiol MEF2C binding had no significant effect. We next tested whether (Figure 7A). However, the presence of exogenous HDAC7 resulted the enzymatic activity of HDAC7 is necessary for its repressive in a significant block in the expression levels of Mac-1 (Figure 7A action on Mac-1 expression during the conversion of pre-B cells and 7B). In contrast, CD19 protein levels decreased to the same into macrophages. We generated a retroviral vector for HDAC7 extent regardless of the presence of HDAC7 (Figure 7A and 7B). mutated in its catalytic domain (HDAC7-H657A), a C-terminal Exogenous expression of HDAC7 in primary B cell precursors also truncated construct HDAC7(1–487) that completely lacks the interfered with the up-regulation of Mac-1 in transdifferentiation HDAC catalytic domain but contains the MEF2 interacting motif experiments (Figure 7C). These results demonstrated that in the and a N-terminal truncated construct HDAC7(438–915) bearing presence of HDAC7 the reprogrammed cells are not able to the enzymatic motif but lacking the MEF2 domain. As shown in express the levels of Mac-1 protein normally present in macro- Figure 8, forced expression of wild-type HDAC7 interfered with phages. Second, we tested the phagocytic properties of the the up-regulation of Mac-1 protein levels 48 hours after induction transdifferentiated C10 cells exogenously expressing HDAC7. of cellular reprogramming. In contrast, the HDAC7-H657A, the Interestingly, C10-HDAC7 cells treated with b-estradiol showed a HDAC7(1–487) and the HDAC7(438–915) constructs did not significantly reduced capacity to phagocytose red fluorescence block the up-regulation of Mac-1 levels. These experiments protein-expressing bacteria compared to the control C10-MSCV demonstrate that both the HDAC7-MEF2C interaction, as well cells (Figure 7D). Lastly, we analyzed the inflammatory response of as its catalytic activity, are necessary for HDAC7 to repress both cell lines after treatment with LPS in the presence or absence macrophage genes during cellular reprogramming. of b-estradiol for 48 hours. Strikingly, expression of HDAC7 resulted in a significant inhibition in the expression of the pro- Discussion inflammatory cytokines, Tnfa, Il-1a and Il-6, by the macrophage- like converted cells in response to LPS (Figure 7E). We next tested Our findings have revealed that HDAC7 is expressed in B cell whether HDAC7 re-expression results in the reduction of Mac-1 precursors and not in macrophages (Figure 9). Using a cellular protein levels once the cells have been reprogrammed into transdifferentiation system we have demonstrated that HDAC7 macrophages. We first treated C10 cells with b-estradiol for represses the expression of a large number of macrophage genes 72 hours and the reprogrammed macrophages were transduced during the conversion of pre-B cells into this myeloid cell type. with a retroviral vector carrying HDAC7-Flag. We observed a More importantly, depletion of HDAC7 in pre-B cells results in decrease in the expression levels of Mac-1 24 hours after the de-repression of key macrophage genes. Functionally, the expression of exogenous HDAC7 (Figure S7A). However, we did presence of HDAC7 interferes with the acquisition of key not observe any effect on either the expression of macrophage functional features of the converted macrophages. These aberrant genes or the phagocytic capacity of RAW cells expressing macrophage-converted cells do not express the Mac-1 protein exogenous HDAC7 (Figure S7B, S7C and S7D). We have recently levels normally present in macrophages, are not able to properly

PLOS Genetics | www.plosgenetics.org 9 May 2013 | Volume 9 | Issue 5 | e1003503

HDAC7 Is a Key Repressor in Pre-B Cells

Figure 6. HDAC7 binds to the transcription factor MEF2C and is recruited to the promoter of key genes for macrophage function in pre-B cells. A. Co-immunoprecipitation experiments showing the specific binding of HDAC7 with MEF2C in pre-B cells. HDAC7 does not bind with IKAROS, PAX5 and E2A. B. Venn diagrams showing the total number of genes up-regulated in MEF2C deficient LMPPs, genes up-regulated at 48 and 72 hours of b-estradiol treatment affected by HDAC7 and the overlapping between both conditions. C and F. Schematic representation of the mouse

PLOS Genetics | www.plosgenetics.org 10 May 2013 | Volume 9 | Issue 5 | e1003503

HDAC7 Is a Key Repressor in Pre-B Cells

Cxcl10 and Fcgr1 locus and amplicons scanned in Chromatin immunoprecipitation experiments by qPCR. Asterisks indicate MEF2 binding sites location. D, E, G and H. Chromatin immunoprecipitation experiments showing the enrichment of HDAC7 and MEF2C to putative MEF2 binding sites on the Cxcl10 and Fcgr1 gene loci in pre-B cells. Results are presented as percentage immunoprecipitated over input and are representative of three independent experiments. doi:10.1371/journal.pgen.1003503.g006 phagocytose bacteria and do not respond adequately to endotoxin HDAC7 becomes phopshorylated and translocates to the by expressing major pro-inflammatory cytokines. At the mecha- cytoplasm where it can no longer repress its target genes nistic level, HDAC7 specifically interacts with the transcription [26,27,29,31]. Later in the periphery, HDAC7 is constitutively factor MEF2C in pre-B cells and is recruited to MEF2 binding phosphorylated and found in the cytoplasm of CTLs allowing the sites located at the promoters of genes critical for macrophage expression of key genes for the function of this specialized T cell function (Figure 6 and Figure S6). In addition to the interaction type [30]. We could postulate that within a particular hemato- with MEF2C, the catalytic activity of HDAC7 is also necessary to poietic cellular lineage, the HDAC7 gene transcriptional repress macrophage genes. Our results demonstrate that HDAC7 repressive function is regulated in a signal-dependent way by is expressed in fetal and adult pre-B cells suggesting that it might controlling its phosphorylation status and altering its cellular play a role in both types of B lymphopoiesis. distribution. That is, HDAC7 is unphosphorylated or phos- Why is HDAC7 specifically expressed in pre-B cells and not phorylated and localized either in the cell nucleus or in the macrophages? Based on our results, we conclude that HDAC7 is a cytoplasm depending on the requirement to silence or activate key transcriptional repressor of lineage inappropriate genes in pre- specific genes at different stages of T cell development. In B cells. Using a ChIP-seq approach Murre and colleagues have contrast, at the branching points where cellular lineage choice identified HDAC7 as a target of the transcription factors E2A, has to be made, e.g. lymphoid versus myeloid, HDAC7 might EBF and Foxo1 in B cell precursors (pro-B cells) postulating that it be regulated at the expression level. For instance, our results could be an important regulator of B cell development and showing that HDAC7 expression is down-regulated during the indicating that HDAC7 function is not restricted to T cells within conversion of pre-B cells into macrophages further support this the hematopoietic system [21]. In addition, our results showing hypothesis. However, as is the case for T cells, we cannot rule that in pre-B cells HDAC7 interacts with the transcription factor out the possibility that HDAC7 is additionally regulated in a MEF2C, allowing the recruitment to the MEF2 binding sites on signal dependent manner during B cell development and promoters of genes characteristic of macrophages reinforces our differentiation. In this regard, we have found that HDAC7 is conclusion. Indeed, in the late nineties, MEF2C was found, among expressed in the nucleus of pre-B cells (data not shown) probably the different MEF2 family members, to be specifically expressed in due to the lack of a fully functional B cell receptor (BCR). B cells within the lymphocyte lineage, suggesting that it could have However, it is highly probable that HDAC7 function is also a role in B cell development and function [22]. More recently, it regulated via BCR signaling at later cell differentiation stages has been reported that MEF2C regulates B cell proliferation and when B cells develop into mature antibody-secreting cells. survival after BCR activation and p38 MAPK signaling [23,24]. Based on the findings of this study, we therefore propose that in MEF2C is also expressed at earlier stages of lymphocyte the hematopoietic system, HDAC7 is not only a signal-dependent development [5,25]. It has been reported that MEF2C is involved transcriptional repressor involved in different developmental steps in the cellular choice towards the lymphoid versus the myeloid of a particular lymphocyte type, but also a lineage-specific lineage in lymphoid-primed multipotent progenitors (LMPPs) transcriptional repressor responsible for maintaining the identity [25]. At the molecular level MEF2C activates the transcription of of lymphocytes by silencing lineage inappropriate genes. We lymphoid specific genes and represses myeloid genes [25]. We anticipate that other members of the class IIa HDACs subfamily have observed a significant overlap between MEF2C regulated might be specific repressors of undesirable genes in the cellular genes in LMPPs and our identified HDAC7 target genes in pre-B differentiation and developmental processes where they exert their cells. Moreover, HDAC7 mutants that do not interact with actions (e.g. skeletal and cardiac muscle, bone formation and MEF2C are unable to suppress the up-regulation of macrophage brain). genes during the reprogramming process. Given this scenario, we propose that HDAC7 is the MEF2C transcriptional co-repressor Materials and Methods responsible for the silencing of myeloid genes in B cells and lymphoid precursors. Plasmids Finally, in light of previous reports demonstrating the expression MSCV-puro-HDAC7, MSCV-GFP-HDAC7, MSCV-GFP- and functions of HDAC7 in T cells, our results have uncovered a HDAC7(DMEF), MSCV-GFP-HDAC7(K86A/K88A), MSCV- more general role for HDAC7 in the lymphoid lineage compart- GFP-HDAC7(H657A), MSCV-GFP-HDAC7(1–487) and ment within the hematopoietic system. In addition, the finding MSCV-GFP-HDAC7(438–915) construct were generated by that HDAC7 is dramatically down-regulated during the transdif- cloning the HDAC7-Flag cDNA PCR amplificated from the ferentiation of pre-B cells into macrophages not only reinforces pCDNA3-HDAC7 wild-type and mutant plasmids previously this notion, but also provides an additional level of complexity to described [20] into the MSCV-puro or MSCV-GFP vectors the way its activity and function are regulated. During T cell (Clontech). development in the thymus and later in a differentiated and specialized T cell type, cytotoxic T lymphocytes (CTLs), HDAC7 Antibodies is regulated in a signal dependent manner responsible for its Anti-HDAC7 (H-273), anti-HDAC7 (C-18), anti-HDAC4 (H- phosphorylation and for its subsequent nucleo-cytoplasmic distri- 92), anti-E2A (V-18), anti-PAX5 (C-20) and anti-RUNX1 (DW71) bution [26–31]. In resting thymocytes, HDAC7 is localized in the were purchased from Santa Cruz Biotechnology. Anti-IKAROS nucleus, where it represses the expression of a large number of (ab26083) was purchased from Abcam; anti-MEF2C (D80C1) XP genes involved in both positive (survival) and negative (apoptosis) from Cell Signaling Technology; and anti-a-Tubulin (T61999), selection of the cells. However, in response to TCR signaling, from Sigma-Aldrich.

PLOS Genetics | www.plosgenetics.org 11 May 2013 | Volume 9 | Issue 5 | e1003503

HDAC7 Is a Key Repressor in Pre-B Cells

Figure 7. HDAC7 re-expression interferes with the functional properties of the converted macrophages. A. Histograms for Mac-1 and CD19 expression levels in C10-MSCV and C10-HDAC7 cells untreated or treated with b-estradiol for 72 hours. B. Mean fluorescence intensity (MFI) of Mac-1 and CD19 of C10-MSCV and C10-HDAC7 cells untreated or treated with b-estradiol for 72 hours. Data are represented as the mean +/2 standard error of the mean (SEM) of three independent experiments. *P,0.001. C. Primary pre-B cells were transduced with the indicated retroviral vectors and 48 hours after induce to transdifferentiate. Histograms for Mac-1 proteins levels are shown. D. Capacity of C10-MSCV and C10-HDAC7 cells untreated or treated with b-estradiol for 48 hours to phagocytose red fluorescence bacteria. Data are given as mean 6 SEM of values obtained

PLOS Genetics | www.plosgenetics.org 12 May 2013 | Volume 9 | Issue 5 | e1003503

HDAC7 Is a Key Repressor in Pre-B Cells

in three independent experiments. Statistical significance was determined by two-way ANOVA followed by Bonferrony multiple comparison test. *P,0.001. E. Effect of HDAC7 expression in LPS-mediated Tnfa, Il-1a and Il-6 gene expression. C10-MSCV and C10-HDAC7 cells were treated or not with b-estradiol for 48 hours. Then, the cells where incuated or not with LPS for 6 hours and RNAs analyzed by RT-qPCR. Data are represented as the mean +/2 standard error of the mean (SEM) of three independent experiments. P values were calculated by the two-way ANOVA test. *P,0.001. doi:10.1371/journal.pgen.1003503.g007

Cell culture and b-estradiol treatment estradiol in the presence of 10 nM of IL-3 and 10 nM of mCSF1 HAFTL cells were grown at 37uC in RPMI 1640 supplemented for the indicated periods of time. with 10% fetal bovine serum, 2 mM glutamine, and 50 U/ml streptomycin/penicillin. C10 cells (transduced with a MSCV- Retroviral supernatant generation and cellular GFP-C/EBPa retroviral vector) and C11 cells (transduced with a transduction MSCV-hCD4-C/EBPa retroviral vector) were cultured at 37uCin For retrovirus generation the MSCV-puro, MSCV-puro- RPMI 1640 without phenol red supplemented with 10% of HDAC7, MSCV-GFP, MSCV-GFP-HDAC7, MSCV-GFP- charcoal treated fetal bovine serum, 2 mM glutamine, and 50 U/ HDAC7(DMEF), MSCV-GFP-HDAC7(K86A/K88A), MSCV- ml streptomycin/penicillin. B-cell precursors and macrophages GFP-HDAC7(H657A), MSCV-GFP-HDAC7(1–487) and MSCV- were obtained from mouse bone marrow as described [18]. For GFP-HDAC7(438–915) plasmids were transfected into the pack- transdifferentiation induction, cells were treated with 100 nM b- aged cell line Platinum-E and supernatant were collected at 48–

Figure 8. Interaction of HDAC7 with MEF2C and its catalytic activity are essential for repression of Mac-1. C11 cells were infected with MSCV-GFP, MSCV-GFP-HDAC7, MSCV-GFP-HDAC7(DMEF), MSCV-GFP-HDAC7(K86A/K88A), MSCV-GFP-HDAC7(H657A), MSCV-GFP-HDAC7(1–487) and MSCV-GFP-HDAC7(438–915) viruses and induced with b-estradiol for 48 hours. A. Cells were stained with a Mac-1 antibody and the GFP-positive fractions were gated and the results plotted. B. Percentage of Mac-1 positive cells treated with b-estradiol for 48 hours. Data are given as mean 6 SEM of values obtained in three independent experiments. Statistical significance was determined by two-way ANOVA followed by Bonferrony multiple comparison test. *P,0.0001. C. RT-qPCR experiments for Itgam gene expression changes in cells treated with b-estradiol for 48 hours. Data are given as mean 6 SEM of values obtained in three independent experiments. Statistical significance was determined by two-way ANOVA followed by Bonferrony multiple comparison test. *P,0.001. doi:10.1371/journal.pgen.1003503.g008

PLOS Genetics | www.plosgenetics.org 13 May 2013 | Volume 9 | Issue 5 | e1003503

HDAC7 Is a Key Repressor in Pre-B Cells

Figure 9. Model for HDAC7-mediated transcriptional repression in pre-B cells. HDAC7 is expressed in pre-B cells and not in macrophages. In pre-B cells, HDAC7 specifically interacts with the transcription factor MEF2C and is recruited to promoters of myeloid genes. During transdifferentiation of pre-B cells into macrophages HDAC7 is down-regulated allowing for expression of macrophages specific genes. doi:10.1371/journal.pgen.1003503.g009

72 hours post-transfection. For the generation of C10-MSCV and Kit (AB Applied Biosystems). RT-qPCR were performed in C10-HDAC7 cells, C10 cells were spin infected and 48 hours after triplicate using SYBR Green I Master (Roche). PCR reactions were selected in the presence of 3 ug/ml of puromycin. C11 cells were were run and analyzed using the LightCycler 480 Detection spin infected with MSCV-GFP, MSCV-GFP-HDAC7, MSCV-GFP- System (Roche). Primers sequences upon request. HDAC7(DMEF), MSCV-GFP-HDAC7(K86A/K88A), MSCV- GFP-HDAC7(H657A), MSCV-GFP-HDAC7(1–487) and MSCV- Co-immunoprecipitation and Western blot analysis GFP-HDAC7(438–915) and 48 hours after treated with b-estradiol. Total cellular extracts were prepared in PLB buffer (0.5% Triton X-100, 0.5 mM EDTA, 1 mM DTT in PBS) supplement- siRNA depletion ed with protease inhibitors (Complete, Roche Molecular Bio- Dharmacon siRNA control and on-target smartpools targeting chemicals). Immunoprecipitation, SDS-PAGE and Western blot transcript of the mouse HDAC7 gene were used to knockdown its experiments were performed as previously described [32]. expression HAFTL cells and primary B cell precursors. siRNA were transfected using Lipofectamine RNAiMAX Transfection Flow cytometry Reagent (Invitrogen). mRNA levels were examined by RT-qPCR Cells were un-induced or induced to transdifferentiate. experiments 72 hours after siRNA transfection. 48 hours later, cells were stained with fluorochrome conjugated antibodies against Mac-1 and CD19 (BD Pharmingen). Mac-1 RT–qPCR experiments and CD19 expression were monitored on a Gallios Flow RNA was extracted Trizol extraction (Qiagen) and cDNA Cytometer (Beckman Coulter) and analyzed by FlowJo software synthesyzed using the High Capacity cDNA Reverse Trancription (Tree Star, Inc.).

PLOS Genetics | www.plosgenetics.org 14 May 2013 | Volume 9 | Issue 5 | e1003503

HDAC7 Is a Key Repressor in Pre-B Cells

Phagocytosis assays generation (www.gitools.org). Resulting p-values were adjusted for C10-MSCV and C10-HDAC7 cells were induced or not for multiple testing using the Benjamin and Hochberg’s method of 48 h and subjected to the pHrodo E.coli phagocytosis KiT False Discovery Rate (FDR). (Invitrogen) following the manufacturers protocol. Chomatin immunoprecipitations assays LPS induced inflammatory cytokines ChIP was performed essentially as previously described [32] on C10-MSCV and C10-HDAC7 cells were induced to transdif- C10 cells un-induced or induced for 72 hours. To shear chromatin ferentiate. 48 h after induction they were incubated with LPS to an apparent length of ,500 bp, chromatin was sonicated using (1 mg/mL Sigma) for 6 h. RNA extraction, cDNA synthesis and a BioRuptor sonicator (Cosmo Bio Co., Ltd) with either 40 45-s RT-qPCR were performed as described above. Primers sequences pulses (uninduced cells) or 30 45-s pulses (induced cells) at upon request. maximum setting. Input and immunoprecipitated DNA were subjected to Sybergreen Q PCR cycles with specific primers Microarray experiments (provided upon request). Biological duplicates of C10-MSCV and C10-HDAC7 cells were un-induced or induced to transdifferentiate for 48 and Supporting Information 72 hours (12 samples in total). Total RNA from cultured cells was Dataset S1 Microarray statistics analysis for genes up-regulated extracted by Trizol and then purified. PCR amplified RNAs were during the conversion of pre-B cells into macrophages that are hybridized against Affymetrix mouse arrays chip (Mouse Genome affected by HDAC7. 430 PM strip) at the IRB Genomics Facility. Affymetrix raw CEL (XLS) files and processed (normalized) data have been deposited in GEO database under accession number GSE36827. Dataset S2 Microarray statistics analysis for genes down- regulated during the conversion of pre-B cells into macrophages Microarray data analysis that are affected by HDAC7. Affymetrix CEL files were background corrected, normalized (XLS) using Bioconductor, package ‘‘affy’’ (version 1.28.1) using ‘expres- Figure S1 HDAC7 is down-regulated during the transifferentia- so’ algorithm [33,34]. Since the Affymetrix chip version used in tion of pre-B cells into macrophages. Kinetics of regulation (log2 this study contains only perfect match (pm) probes, for Affymetrix expression values) of Hdac7 and Pax5 genes in C10 cells normalization and acquiring raw probe intensities to expression and primary pre-B cells transduced with a retroviral vector for values we used the following parameters: background correction inducible expression of C/EBPa. Cells were treated with b- method ‘‘rma’’; normalization method ‘‘constant’’; pm correct estradiol for the times indicated. HAFTL cells, primary pre-B cells, method ‘‘pmonly’’; and summary method ‘‘avgdiff’’. Quality of primary bone marrow macrophages and RAW264.7 cells were microarray experiment (data not shown) was verified by used as control. Bioconductor package ‘‘arrayQualityMetrics’’ (version 3.2.4 (EPS) under Bioconductor version 2.7; R version 2.12.1) [35]. To determine genes that are differentially expressed (DE) between Figure S2 Western blot showing the protein levels of endoge- two experimental conditions, Bioconductor package Limma was nous and exogenously expressed HDAC7 in C10 cells transduced utilized to generate contrast matrices and fit the corresponding with a untreated or treated with b-estradiol for 72 hours. linear model [36]. Probe to gene annotation were performed (EPS) using microarray vendor’s annotation data. When more than Figure S3 HDAC7 re-expression interferes with the gene one probe were annotated to same gene, highest absolute transcriptional program of the converted macrophages. A. Heat- expression value was considered (maximizing). To consider a map statistics showing significantly (FDR p-value #0.05) enriched gene is differentially expressed, besides multiple test corrected, KEGG pathways among the up-regulated genes affected by the re- FDR p-value #0.05 as cut off, we also applied Log2 fold change expression of HDAC7 during transdifferentiation of pre-B cells (Log2FC) cut off 0.5 for b-estradiol treatment. We used Log2FC into macrophages. Colours toward red indicate high statistic cut off 0.5 for genes that are affected by the expression of significance, yellow indicates low statistic significance, and gray HDAC7 in b-estradiol treated cells. Expression data on Mef2c indicates no statistic significance. B. Heat-maps showing observed deficient multipotent progenitor cells were obtained from GEO differentially expressed genes for selected KEGG pathways. Blue database (accession No. GSE13686) [5]. Data were analyzed colour cell indicates positive events while gray colour indicates that using the limma package from Bioconductor. Spots were not the gene was not observed differentially expressed in that background corrected before within array loess normalization. experimental condition. After array normalization using the quantile method log2 ratios (mutant/control) was calculated. To define a gene up-regulated, (EPS) we used Log2FC $1.0. Figure S4 HDAC7 re-expression interferes with the gene transcriptional program of the converted macrophages. Heatmap Functional and pathway enrichment analysis statistics showing significantly (FDR p-value #0.05) enriched A. Functional annotation of differentially expressed genes is based GO Biological Processes categories, B. GO Cellular Components on Gene Ontology (GO) (http://www.geneontology.org) as categories, C. GO Molecular Functions categories and D. KEGG extracted from EnsEMBL and KEGG pathway database. pathways, among the down-regulated genes affected by the re- Accordingly, all genes are classified into three ontology categories expression of HDAC7 during transdifferentiation of pre-B cells (i) biological process (BP), (ii) cellular component, (CC) and into macrophages. Colours toward red indicate high statistic molecular function (MF) and pathways when possible. We have significance, yellow indicates low statistic significance, and gray taken only the GO/pathway categories that have at least 10 genes indicates no statistic significance. annotated. We used Gitools for enrichment analysis and heatmap (EPS)

PLOS Genetics | www.plosgenetics.org 15 May 2013 | Volume 9 | Issue 5 | e1003503

HDAC7 Is a Key Repressor in Pre-B Cells

Figure S5 HDAC7 re-expression does not interfere with the either an empty vector or a retroviral vector for HDAC7 down-regulation of Pax5. A. Kinetics of down-regulation (log2 expression. B. Histograms for Mac-1 and CD19 protein levels in Affymetrix expression values) of Pax5 in C10-MSCV and C10- RAW-MSCV and RAW-HDAC7 cells. C. RT-qPCR experi- HDAC7 cells un-treated or treated with b-estradiol for the ments for gene expression changes for Hdac7, Itgam, Ccl3, and times indicated. B. RT-qPCR validation of the results shown Fcgr1 genes in RAW-MSCV and RAW-HDAC7 cells. D. in A. Capacity of RAW-MSCV and RAW-HDAC7 cells to phagocytose (EPS) red fluorescence bacteria. Figure S6 HDAC7 is recruited to the Itgam promoter in pre-B (EPS) cells. A. Schematic representation of the mouse Itgam locus and amplicons scanned in Chromatin immunoprecipitation experi- Aknowledgments ments by qPCR. Asterisks indicate MEF2 binding sites location. B. We thank Dr. Eric Verdin for facilitating the pcDNA3.1-HDAC7 Chromatin immunoprecipitation experiments showing the enrich- constructs. We thank Dr. Pura Munoz-Canoves, Dr. Esteban Ballestar, ment of HDAC7 and MEF2C to putative MEF2 binding sites on and Dr. Alejandro Vaquero for helpful comments on the manuscript and the Itgam gene loci in pre-B cells. Results are presented as Dr. Fernando Setien for technical advice. percentage immunoprecipitated over input and are representative of three independent experiments. Author Contributions (EPS) Conceived and designed the experiments: MP. Performed the experiments: Figure S7 HDAC7 re-expression decreases Mac1 protein levels BB-Z LR-G OC HR ABMMKI TM. Analyzed the data: BB-Z LR-G OC in the reprogrammed macrophages. A. Histograms for Mac-1 HR ABMMKI TM NL-B MP. Contributed reagents/materials/analysis protein levels in reprogrammed machrophages transduced with tools: LHB AdT LDA TG NL-B. Wrote the paper: MP.

References 1. Parra M (2009) Epigenetic events during B lymphocyte development. 20. Perez-Llamas C, Lopez-Bigas N (2011) Gitools: analysis and visualisation of Epigenetics 4: 462–468. genomic data using interactive heat-maps. PLoS ONE 6: e19541. doi:10.1371/ 2. Mercer EM, Lin YC, Murre C (2011) Factors and networks that underpin early journal.pone.0019541 hematopoiesis. Semin Immunol 23: 317–325. 21. Lin YC, Jhunjhunwala S, Benner C, Heinz S, Welinder E, et al. (2010) A global 3. Adolfsson J, Mansson R, Buza-Vidas N, Hultquist A, Liuba K, et al. (2005) network of transcription factors, involving E2A, EBF1 and Foxo1, that Identification of Flt3+ lympho-myeloid stem cells lacking erythro-megakaryo- orchestrates B cell fate. Nat Immunol 11: 635–643. cytic potential a revised road map for adult blood lineage commitment. Cell 121: 22. Swanson BJ, Jack HM, Lyons GE (1998) Characterization of myocyte enhancer 295–306. factor 2 (MEF2) expression in B and T cells: MEF2C is a B cell-restricted 4. Yoshida T, Ng SY, Zuniga-Pflucker JC, Georgopoulos K (2006) Early transcription factor in lymphocytes. Mol Immunol 35: 445–458. hematopoietic lineage restrictions directed by Ikaros. Nat Immunol 7: 382–391. 23. Wilker PR, Kohyama M, Sandau MM, Albring JC, Nakagawa O, et al. (2008) 5. Stehling-Sun S, Dade J, Nutt SL, DeKoter RP, Camargo FD (2009) Regulation Transcription factor Mef2c is required for B cell proliferation and survival after of lymphoid versus myeloid fate ‘choice’ by the transcription factor Mef2c. Nat antigen receptor stimulation. Nat Immunol 9: 603–612. Immunol 10: 289–296. 24. Khiem D, Cyster JG, Schwarz JJ, Black BL (2008) A p38 MAPK-MEF2C 6. Bain G, Maandag EC, Izon DJ, Amsen D, Kruisbeek AM, et al. (1994) E2A pathway regulates B-cell proliferation. Proc Natl Acad Sci U S A 105: 17067– proteins are required for proper B cell development and initiation of 17072. immunoglobulin gene rearrangements. Cell 79: 885–892. 25. Gekas C, Rhodes KE, Gereige LM, Helgadottir H, Ferrari R, et al. (2009) 7. Bain G, Robanus Maandag EC, te Riele HP, Feeney AJ, Sheehy A, et al. (1997) Mef2C is a lineage-restricted target of Scl/Tal1 and regulates megakaryopoiesis Both E12 and E47 allow commitment to the B cell lineage. Immunity 6: 145– and B-cell homeostasis. Blood 113: 3461–3471. 154. 26. Dequiedt F, Kasler H, Fischle W, Kiermer V, Weinstein M, et al. (2003) 8. Zhuang Y, Soriano P, Weintraub H (1994) The helix-loop-helix gene E2A is HDAC7, a thymus-specific class II histone deacetylase, regulates Nur77 required for B cell formation. Cell 79: 875–884. transcription and TCR-mediated apoptosis. Immunity 18: 687–698. 9. Lin H, Grosschedl R (1995) Failure of B-cell differentiation in mice lacking the 27. Kasler HG, Verdin E (2007) Histone deacetylase 7 functions as a key regulator transcription factor EBF. Nature 376: 263–267. of genes involved in both positive and negative selection of thymocytes. Mol Cell 10. Dengler HS, Baracho GV, Omori SA, Bruckner S, Arden KC, et al. (2008) Biol 27: 5184–5200. Distinct functions for the transcription factor Foxo1 at various stages of B cell 28. Kasler HG, Young BD, Mottet D, Lim HW, Collins AM, et al. (2011) Histone differentiation. Nat Immunol 9: 1388–1398. deacetylase 7 regulates cell survival and TCR signaling in CD4/CD8 double- 11. Urbanek P, Wang ZQ, Fetka I, Wagner EF, Busslinger M (1994) Complete positive thymocytes. J Immunol 186: 4782–4793. block of early B cell differentiation and altered patterning of the posterior 29. Dequiedt F, Van Lint J, Lecomte E, Van Duppen V, Seufferlein T, et al. (2005) midbrain in mice lacking Pax5/BSAP. Cell 79: 901–912. 12. Delogu A, Schebesta A, Sun Q, Aschenbrenner K, Perlot T, et al. (2006) Gene Phosphorylation of histone deacetylase 7 by protein kinase D mediates T cell repression by Pax5 in B cells is essential for blood cell homeostasis and is receptor-induced Nur77 expression and apoptosis. J Exp Med 201: 793–804. reversed in plasma cells. Immunity 24: 269–281. 30. Navarro MN, Goebel J, Feijoo-Carnero C, Morrice N, Cantrell DA (2011) 13. Schebesta A, McManus S, Salvagiotto G, Delogu A, Busslinger GA, et al. (2007) Phosphoproteomic analysis reveals an intrinsic pathway for the regulation of Transcription factor Pax5 activates the chromatin of key genes involved in B cell histone deacetylase 7 that controls the function of cytotoxic T lymphocytes. Nat signaling, adhesion, migration, and immune function. Immunity 27: 49–63. Immunol 12: 352–361. 14. Pridans C, Holmes ML, Polli M, Wettenhall JM, Dakic A, et al. (2008) 31. Parra M, Kasler H, McKinsey TA, Olson EN, Verdin E (2005) Protein kinase Identification of Pax5 target genes in early B cell differentiation. J Immunol 180: D1 phosphorylates HDAC7 and induces its nuclear export after T-cell receptor 1719–1728. activation. J Biol Chem 280: 13762–13770. 15. Parra M, Verdin E (2010) Regulatory signal transduction pathways for class IIa 32. Rafati H, Parra M, Hakre S, Moshkin Y, Verdin E, et al. (2011) Repressive LTR histone deacetylases. Curr Opin Pharmacol 10: 454–460. nucleosome positioning by the BAF complex is required for HIV latency. PLoS 16. Martin M, Kettmann R, Dequiedt F (2009) Class IIa histone deacetylases: Biol 9: e1001206. doi:10.1371/journal.pbio.1001206 conducting development and differentiation. Int J Dev Biol 53: 291–301. 33. Gentleman RC, Carey VJ, Bates DM, Bolstad B, Dettling M, et al. (2004) 17. Bussmann LH, Schubert A, Vu Manh TP, De Andres L, Desbordes SC, et al. Bioconductor: open software development for computational biology and (2009) A robust and highly efficient immune cell reprogramming system. Cell bioinformatics. Genome Biol 5: R80. Stem Cell 5: 554–566. 34. Gautier L, Cope L, Bolstad BM, Irizarry RA (2004) affy–analysis of Affymetrix 18. Di Tullio A, Manh TP, Schubert A, Mansson R, Graf T (2011) CCAAT/ GeneChip data at the probe level. Bioinformatics 20: 307–315. enhancer binding protein alpha (C/EBP(alpha))-induced transdifferentiation of 35. Kauffmann A, Gentleman R, Huber W (2009) arrayQualityMetrics–a pre-B cells into macrophages involves no overt retrodifferentiation. Proc Natl bioconductor package for quality assessment of microarray data. Bioinformatics Acad Sci U S A 108: 17016–17021. 25: 415–416. 19. Rodriguez-Ubreva J, Ciudad L, Gomez-Cabrero D, Parra M, Bussmann LH, et 36. Smyth GK (2004) Linear models and empirical bayes methods for assessing al. (2011) Pre-B cell to macrophage transdifferentiation without significant differential expression in microarray experiments. Stat Appl Genet Mol Biol 3: promoter DNA methylation changes. Nucleic Acids Res. Article3.

PLOS Genetics | www.plosgenetics.org 16 May 2013 | Volume 9 | Issue 5 | e1003503

Hindawi Publishing Corporation Comparative and Functional Genomics Volume 2012, Article ID 564381, 10 pages doi:10.1155/2012/564381

Review Article Epigenetic Regulation of B Lymphocyte Differentiation, Transdifferentiation, and Reprogramming

Bruna Barneda-Zahonero,1 Lidia Roman-Gonzalez,1 Olga Collazo,1 Tokameh Mahmoudi,2 and Maribel Parra1

1 Cellular Differentiation Group, Cancer Epigenetics and Biology Program (PEBC), Bellvitge Biomedical Research Institute (IDIBELL), Avenida Gran Via s/n km 2.7, 08907 L’Hospitalet de Llobregat, Barcelona, Spain 2 Department of Biochemistry, Erasmus University Medical Center, Ee622, P.O.Box 2040, 3000 CA Rotterdam, The Netherlands

Correspondence should be addressed to Maribel Parra, [email protected]

Received 7 February 2012; Revised 21 May 2012; Accepted 6 July 2012

Academic Editor: Sonia Vanina Forcales

Copyright © 2012 Bruna Barneda-Zahonero et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

B cell development is a multistep process that is tightly regulated at the transcriptional level. In recent years, investigators have shed light on the transcription factor networks involved in all the differentiation steps comprising B lymphopoiesis. The interplay between transcription factors and the epigenetic machinery involved in establishing the correct genomic landscape characteristic of each cellular state is beginning to be dissected. The participation of “epigenetic regulator-transcription factor” complexes is also crucial for directing cells during reprogramming into pluripotency or lineage conversion. In this context, greater knowledge of epigenetic regulation during B cell development, transdifferentiation, and reprogramming will enable us to understand better how epigenetics can control cell lineage commitment and identity. Herein, we review the current knowledge about the epigenetic events that contribute to B cell development and reprogramming.

1. Introduction Every step in B cell development is characterized by the activation of the specific genetic program characteristic of Hematopoietic stem cells (HSCs) give rise to mature B cells the new intermediate/progenitor generated and the repres- through the sequential differentiation of lymphoid pro- sion/extinction of the genetic program of the previous genitor cells. Long-term HSCs (LT-HSCs) have the ability cellular state. To achieve this, the different differentiation to self-renew and reconstitute the entire immune system by differentiating into short-term HSCs (ST-HSCs). ST- steps are tightly regulated at the transcriptional level. In HSCs differentiate into multipotent progenitors (MPPs) that recent years, the theory of the existence of networks of then branch into common myeloid progenitors (CMPs) lineage-specific and identity-transcription factors responsi- and lymphoid-primed multipotent progenitors (LMPPs). ble for establishing particular genomic landscapes has gained CMPs further differentiate into erythrocytes and megakary- credence [6]. In the case of lymphocyte development, the ocytes, whereas LMPPs retain the capability to give rise to transcription factors Ikaros and PU.1 are critical for the myelomonocytic or lymphoid lineages [1, 2]. LMPPs become cellular commitment of LMPPs to the lymphoid lineage common lymphoid progenitors (CLPs) [3],whichhavethe [2]. Subsequently, early B cell specification depends on the potential to differentiate into B and T lymphocytes as well action of E2A, EBF, and FOXO1, whereas Pax5 is required as natural killer (NK) cells [4, 5]. Once committed to the for proper B cell development and for maintaining B cell lymphoid lineage, further differentiation steps lead to the identity [7–12]. Finally, during later developmental stages, formation of pro-B and pre-B cells, which are the early B cell the transcriptional repressors Bcl6 and Blimp-1 are crucial precursors for immature B cells, the terminally differentiated for the generation of germinal-center B cells and plasma cells, plasma cells and germinal-center B cells (Figure 1). respectively [13–17](Figure 1).

2 Comparative and Functional Genomics

miR-150 c-Myb miR-181 miR-17-92 miR-34a miR-185 miR-155 Pten Foxp1 and Bim BCR Spfi1 and Aicda Ago2 Dicer Dicer Pre-BCR BCR

LMPP CLP Pro-B Pre-B Immature Mature GC Plasma B cell B cell B cell cell

AID PRC2 G9a IRF4 HDACs Bach2 Bcl6 Blimp-1 Prmt5 PU.1 IKAROS E2A EBF FOXO1 Pax5 Mi-2/NuRD B cell identity LSD1 MYSM1 IRF8 mb-1; cd19; c-fms... HDACs Irf4 Linker histone 1 Mi-2/NuRD Xbp1 SWI/SNF

Figure 1: Scheme for B cell development. Successive stages of B cell differentiation and the key transcription factors and epigenetic regulators involved are shown. The epigenetic regulators that cooperate with specific transcription factors at every cell differentiation step are in purple. MicroRNA transcript targets are in green.

The picture of the hierarchical network of transcription sites. Second, DNA methylation results in the recruitment of factors that mediate the epigenetic signature needed to methyl-CpG-binding proteins (MeCPs and MBDs) in asso- regulate the specific transcriptome of the fate of B-cells ciation with corepressor complexes. Both mechanisms lead during their development has begun to emerge [18–20]. For to the transcriptional silencing of the methylated genes [29]. example, Pax5, whose expression is induced by E2A and The posttranslational modification of histones is another EBF, recruits chromatin-remodeling, histone-modifying and important epigenetic regulatory mechanism. Histones can transcription-factor complexes to its target genes to activate be posttranslationally modified by a variety of enzymatic the transcription of B cell-specific genes, and to silence modifications, including acetylation, methylation, phospho- lineage-inappropriate genes [19]. Extensive efforts have been rylation, sumoylation, and ubiquitination among others made to elucidate the epigenetic mechanisms underlying [30]. While acetylation is generally considered to be a mark the gene rearrangements of various components of the B of transcriptional activation, histone methylation can result cell receptor (BCR) [21–23]. Thus, epigenetic regulation in either transcriptional activation or repression, depending is a critical event in B lymphocyte development. The on the residue that is modified. In this regard, acetylation of relevance of transcription factors to the establishment and histone H3 on lysine 9, 14, and or 18 (H3K9ac, H3K14ac, maintenance of cell-lineage identity has also been demon- H3K18ac) is associated with transcriptional activation and strated in cellular reprogramming experiments [24–27]. The considered “histone active marks.” In the case of histone epigenetic mechanisms involved in the reprogramming and methylation, di- and tri-methylation of histone H3 at lysine transdifferentiation of B cells have also been a focus of study 4 (H3K4me2, H3K4me3) are associated with transcriptional in recent years. activation and therefore considered an active mark, whereas Nucleosomes are the basic unit of the chromatin. They trimethylation of H3 at lysine 27 (H3K27me3) is found to be comprise 147 bp of DNA wrapped around a histone core, enriched at silenced genes and considered to be a repressive which contains two copies each of H2A, H2B, H3, and histone mark [28, 30]. Trimethylation of histone H3 at lysine H4. This core is important for establishing interactions 9 (H3K9me3) has also been characterized as a mark of between nucleosomes and within the nucleosome itself transcriptional repression [30]. However, different reports [28]. Depending on the epigenetic modifications on the suggest that it can also represent transcriptional activity [31, histone tails and in the DNA, chromatin can adopt different 32]. Another mechanism of epigenetic regulation involves structural conformations that are correlated with its active, the action of chromatin remodelers, which are multi-subunit permissive (primed), or repressive status. The four main complexes that use the energy from ATP hydrolysis to change mechanisms by which epigenetic regulation occurs are DNA the location or conformation of nucleosomes, resulting in methylation, histone modification, chromatin remodeling, increased or decreased DNA accessibility [28]. Chromatin- and regulation of gene expression by the action of noncoding remodeling complexes can be divided into four groups, RNAs. The methylation of cytosine residues at CpG dinu- characterized by core ATPase subunits. Based on the defining cleotides (methyl-CpG), which is generally associated with ATPase, they are referred to as the SWI/SNF, ISWI, CHD, transcriptional repression, is accomplished via the action and INO80 families of remodelers [28]. Finally, microRNAs of DNA methyltransferases (DNMTs) [29]. Methyl-CpG- (miRNAs), a type of small noncoding RNAs, have been mediated transcriptional repression can be explained by shown to anneal to 3UTR of cognate mRNAs, leading two nonmutually exclusive molecular mechanisms. First, to mRNA instability and/or the inhibition of translation, methylation of DNA can interfere with the accessibility and thereby making it possible to modulate the proteome of the recruitment of transcription factors to their DNA-binding cell [29].

Comparative and Functional Genomics 3

In this papre we will summarize the recent advances in chain, IgH and IgL, respectively. The expression of the BCR our understanding of the epigenetic mechanisms controlling subunits VpreB, λ5,andmb-1 (Cd79a), and the initiation B cell development and reprogramming. of D-J rearrangements at the IgH locus defines early B cell commitment [41]. The specification of CLPs in the B cell lineage requires two transcription factors, E2A and EBF1, 2. B Cell Development: Early Specification which have been shown to activate the expression of genes towards the Lymphoid Lineage essential for the formation of pro-B cells [42]. E2A and EBF knockout mouse models are phenotypically similar, When cells are at the LMPP stage, two transcription factors, and both transcription factors are considered to play key Ikaros and PU.1, play critical roles in the early cellular spec- roles in initiating B lymphopoiesis. E2A-deficient mice show ification towards the lymphoid lineage. Mice homozygous arrested B cell development at the pre-pro-B cell stage with for a germline mutation in the Ikaros DNA-binding domain compromised D-J rearrangements at the IgH locus and a lack present a block at early lymphocyte development and of expression of Rag1, mb-1, Iν, λ5, Cd19,andPax5 genes therefore lack lymphocyte progenitors, T and B lymphocytes, [7–9]. More recently, it was shown that conditional deletion as well as natural killer cells [33, 34]. More recently, Ikaros of E2A in pre-B cells did not result in a complete loss of was shown to be a crucial transcription factor for the expression of its target genes, indicating the involvement of commitment of LMPPs into CLPs, clearly demonstrating its E2A in the early steps of B cell commitment [43]. Similar to key role in the early cellular decision to undergo lymphocyte E2A, EBF is also known to play a crucial role in initiating development [2]. LMPPs derived from Ikaros-null mice lack B cell development. Mice lacking EBF do not express Rag1, B cell potential and do not express Flt3, Il-7r, Rag1 and Rag2, mb-1, B29 (Igβ), λ5, VpreB, cd19,orPax5 genes [10]. Rag2, which are important genes for lymphoid commitment Recent studies have also implicated the transcription factor [2]. Mechanistically, Ikaros can either activate or repress FOXO-1 in early B lymphopoiesis. FOXO-1-deficient mice transcription of target genes, depending on the recruitment also show a developmental block at the pro-B cell stage [11]. of coactivators or corepressors. For example, in T cells, Moreover, it has been reported that FOXO-1 regulates Rag1 Ikaros has been shown to recruit corepressor or chromatin and Rag2 expression [44]. remodeling complexes in order to either repress or activate Recent evidence indicates that the network of tran- specific targets [35–37]. However, how Ikaros mediates scription factors Pax5, E2A and EBF also cooperate to the epigenetic regulation of its target genes during the regulate their target genes. For example, E2A, EBF, and Pax5 differentiation of LMPPs into CLPs remains to be elucidated. coordinate epigenetic events that lead to the expression of Likewise, the transcription factor PU.1 is crucial for the mb-1, which encodes the Igα subunit of the pre-BCR and commitment of LMPPs to the lymphoid lineage. Strikingly, BCR [45]. mb-1 is methylated at CpG dinucleotides in HSCs PU.1 is also required for the generation of GMPs and and is gradually demethylated during B cell commitment macrophages. In fact, mice deficient for PU.1 die around correlating with its pattern of expression [21]. EBF and birth and lack B, T, NK and myelomonocytic cells [38, 39]. E2A contribute to the CpG demethylation and nucleosomal The promiscuity of PU.1 in regulating gene expression in remodeling of the mb-1 promoter, an event necessary different cell types raised the general question of what the for its transcriptional activation by Pax5. ATP-dependent mechanism of action is of a given transcription factor in chromatin remodeling complexes have also been implicated different cell types. In this regard, Heinz et al. recently in EBF and Pax5-mediated regulation of the mb-1 gene identified the genomewide binding sites of PU.1 in splenic B [21]. Knockdown of Brg1 and Brm, the catalytic subunits cells, macrophages and B cell progenitors [40]. They found of the SWI/SNF chromatin-remodeling complex interfere that PU.1 cooperates with cell-type-specific transcription with EBF and Pax5-mediated activation of mb-1. In contrast, factors to activate the cisregulatory elements required for knockdown of Mi-2, the catalytic subunit of the Mi-2/NuRD the development of a particular cell type. For example, in chromatin-remodeling complex, enhances chromatin acces- CLPs and pro-B cells, E2A induces PU.1 binding at B cell- sibility and demethylation of the mb-1 promoter and its specific genomic sites that contain closely located PU.1 and transcription in response to both transcription factors [21]. E2A binding motifs [40]. In addition, PU.1 binding initiates These results are consistent with a model in which the nucleosome remodeling, followed by H3K4me enrichment at SWI/SNF and Mi-2/NuRD chromatin remodeling complexes many specific genomic regions [40]. These data could lead us play antagonistic regulatory roles to enable or limit the to speculate that cooperation between PU.1 and Ikaros might reprogramming of target genes by EBF and Pax5 during be crucial for the activation of specific genes required to B cell development [21]. The B-cell-specific gene Cd19 is specify LMPP into CLPs. Also, the identity of Ikaros and PU.1 another example of a gene that is epigenetically regulated epigenetic partners remains unknown. This matter awaits during early B cell development. Cd19 encodes a cell surface investigation. protein that participates in signal transduction mechanisms via the BCR and pre-BCR. Chromatin remodeling at the 3. B Cell Development: Early upstream enhancer sequences of Cd19 occurs in multipotent B Cell Commitment progenitors [22]. This chromatin remodeling has been shown to facilitate the recruitment of E2A to this locus B cell development is characterized by the generation of the followedbyEBFandPax5recruitment[22]. Interestingly, the BCR, which consists of a heavy and a light immunoglobulin Cd19 promoter is transcriptionally activated only after Pax5

4 Comparative and Functional Genomics binding. In this context, Mercer et al. recently reported that mice resulted in an increase in the number of B cells [51]. the monomethylation of H3K4 (H3K4me) at the enhancer Thus, miR-181 appears to target and repress the transcripts regions of cell lineage-specifying genes is the main epigenetic of critical genes involved in generating B cells. A similar mark, which is associated with their specific expression experimental approach was used to show that another pattern throughout the lymphoid differentiation program microRNA, miR-150, which is expressed in mature B and [46]. Taken together, these reports provide clear examples of T cells, can block B cell differentiation at the pro-B cell how B cell lineage-specific transcription factors cooperatively stage when expressed prematurely [52]. Accordingly, the mediate the epigenetic regulation of target genes during B laboratory of Klaus Rajewsky reported that miR-150 plays lymphopoiesis. a role during B cell differentiation through its action on c- The recent advances in ultrasequencing technologies are Myb expression [53]. Other miRNAs have been associated helping to draw a global picture of how the networks of with the early development of B cells. For instance, miR- transcription factors modify the chromatin of their target 34a ablation results in a developmental block at the pre- genes. The laboratory of Cornelis Murre, using a ChIP-seq B cell stage, and miR-17-92 knockout mice exhibit a block experimental approach, has elucidated how the network of in pro-B cells [54, 55]. They regulate the Foxbp1 and Bim transcription factors E2A, EBF and FOXO-1 orchestrates B and PTEN genes, respectively, which are known to have a cell commitment [18]. They found that during the transition roleinBcelldifferentiation [54, 55]. Recently, Kuchen et of pre-pro-B cell to pro-B cells, E2A-associated genes become al. have elucidated the microRNAome during lymphopoiesis monomethylated at lysine 4 on H3 (H3K4me), a mark at the genome-wide scale, leading to the identification of mainly found on gene enhancer elements. Subsequently, EBF miRNAs that are primed for expression at different stages of and FOXO1 are involved in the enrichment of active histone differentiation [56]. They reported that miRNA expression is modifications such as H3K4me3 on B-cell-specifying genes, tightly regulated by epigenetic modifications. In particular, such as Pax5 [18]. Recently, Treiber and colleagues have they showed that the repressive mark H3K27me3 is associ- shed light on the EBF-mediated epigenetic regulation of its ated with the gene silencing of lineage-inappropriate miRNA target genes [47]. They classified EBF targets as activated, during lymphopoiesis [56]. However, they also observed that repressed, or primed genes. They observed that, in pro-B and active epigenetic regulation by the presence of H3K4me also pre-B cells, the “activated” genes are enriched in H3K4me3 occurs in some of the microRNAs “primed” to be expressed. and H3 acetylation active marks and show low levels of the On the basis of the restrictive expression and abundance of repressive mark H3K27me3 [47]. In contrast, the “repressed” miRNAs during B cell lineage specification, miR-320, miR- genes show the opposite pattern of histone modifications. 191, miR-139 and miR28 appear to be potential regulators The “primed” genes are enriched in the gene enhancer mark of B cell differentiation [56]. The transcripts targeted by key H3K4me in pre-B and pro-B cells and enriched in H3K4me3 miRNAs for the early differentiation of B cells remains to be and H3 acetylation in mature B cells [47]. The identification identified. of the epigenetic regulators recruited by transcription factors to mediate gene expression changes during B lymphopoiesis 4. B Cell Development: Pax5 in the Maintenance remains to be addressed. of B Cell Identity Other epigenetic marks, such as ubiquitination of His- tone H2A, have proved to play a role in early B cell The transcription factor Pax5 is essential for maintaining the development. Jiang et al. pointed out that the histone H2A fate of B cells and is therefore considered to be “the guardian deubiquitinase MYSM1 is an important factor in B cell devel- of B cell identity” [57]. Its expression gradually increases in a opment [48]. Mysm1 knockout mice show a drastic decrease stepwise manner during B cell development. Pax5 expression in the number of B cells in the bone marrow, peripheral is first detected at the early pro-B cell stage and maintained blood, and lymph nodes [48]. The authors concluded that up to the mature B cell stage. Pax5 knockout mice show MYSM1 antagonizes the action of the polycomb repressive a block in B cell development at the pro-B stage [12]. complex 1 (PRC1) on the Ebf1 promoter, enabling lineage- Pax5−/− pro-B cells express both E2A and EBF transcription specific transcription factors, such as E2A, to be recruited to factors, as well as their target genes. In contrast, E2A−/− the Ebf1 locus and to induce its transcription [48]. and EBF−/− derived cells do not express Pax5. Collectively, Early B cell development is also known to be regulated these data indicate that Pax5 is a target for both transcription, by microRNAs. Mice deficient in Ago2, which encodes a factors. The laboratory of Meinrad Busslinger has shed protein essential for microRNA biogenesis and function, light on the molecular mechanisms involved in the gradual display a block in B cell development at the pro-B cell stage expression of Pax5 during B cell development. In particular, [49]. Consistent with this, specific deletion of Dicer in pro- they have identified an enhancer in the Pax5 locus, which in B cells, which abolishes the entire miRNA network in B combination with the promoter, recapitulates B lymphoid cells, results in a complete block of B cell differentiation Pax5 expression [58]. Interestingly, the Pax5 enhancer is at the transition from pro-B to pre-B-cells [50]. Another silenced by DNA methylation in embryonic stem cells, study reported that miR-181, one of the approximately 100 while it becomes activated in multipotent hematopoietic microRNAs known to be expressed in mouse bone marrow progenitors. The presence of consensus binding sites for the cells, is more abundant in the B cell lineage than in other cell transcription factors PU.1, IRF4, IRF8, and NF-k B within types [51]. Transplantation of multipotent hematopoietic the Pax5 enhancer suggests that these transcription factors progenitors overexpressing miR-181 into lethally irradiated play a role in sequential enhancer activation in hematopoietic

Comparative and Functional Genomics 5 progenitors and during B cell development [58]. At the onset methylation as a mechanism for Csf1r silencing during B of pro-B cell development the transcription factor EBF1 cell development. Despite being silenced, Csf1r chromatin induces chromatin remodeling at the Pax5 promoter region. remains in a poised or primed conformation even in mature In non-B cells, Polycomb group proteins repress the Pax5 B cell stages. Importantly, Csf1r expression can be reactivated promoter region [58]. by conditional deletion of the transcription factor Pax5 [62]. In addition to the epigenetic regulation of its expression Pax5 was shown to bind the Csf1r gene directly, resulting during B cell development, Pax5 induces the establishment in loss of RNA polymerase II recruitment and binding of of a B cell-specific transcription program that is associated myeloid transcription factors at cisregulatory elements [63]. with the suppression of inappropriate genes of alternative Finally, Pax5 in conjunction with linker histone H1 also lineages, thereby ensuring its role in maintaining B cell iden- coordinates DNA methylation and histone modifications in tity and differentiation. Using gene expression microarrays the 3 regulatory region of the immunoglobulin heavy chain and genome-wide ChIP-on-chip experimental approaches, locus and thus epigenetically regulates the IgH locus [64]. the laboratories of Busslinger and Nutt have described the complex gene regulatory network regulated by Pax5 during 5. B Cell Development: Terminal Blymphopoiesis[59–61]. These studies have identified genes Differentiation that are activated or repressed by Pax5 in wildtype pro- B cells. Pax5-activated genes appear to encode transcrip- The completion of V(D)J recombination and expression of tion factors and key proteins involved in B cell signaling, the BCR on the surface of B cells marks the beginning of adhesion, migration, antigen presentation and germinal- antigen-dependent B cell development. From this point, B center B cell formation [59, 61]. However, Pax5-repressed cells undergo terminal differentiation dependent on signals genes encode secrete proteins, cell adhesion molecules, emanating from the BCR after antigen triggering [65]. signal transducers and nuclear proteins that are specific Peripheral B cells, without antigen-mediated signaling, are to erythroid, myeloid, and T cell lineages [59, 61]. Pax5- in a resting state [66]. Once activated, they either initiate the activated genes in pro-B cells were found to be enriched with germinal center (GC) reaction or differentiate into antibody- epigenetically active marks, including H3K9ac, H3K4me2 secreting plasma cells. Entry into the GC reaction is regulated and H3K4me3 [60]. Importantly, in Pax5-deficient pro-B by Bcl6, whereas the generation of antibody-secreting plasma cells, these active histone marks were dramatically reduced cells is controlled by Blimp-1. Bcl6 and Blimp-1 both act as or lost, indicating that Pax5 is essential for guaranteeing transcriptional repressors and work in a mutually exclusive the active chromatin structure at its target genes. These manner [67, 68]. findings demonstrate that Pax5 is a master regulator of B After antigen triggering, Bcl6 is upregulated in some B cell identity, which, in conjunction with epigenetic regula- cells that then enter the GC reaction [13–15]. In contrast, tors, coordinates a B-cell-specific target gene transcription cells in which Bcl6 is not upregulated undergo differentiation program. Recently, McManus and colleagues have described into plasma cells [69, 70]. From a mechanistic angle, Bcl6 the epigenetic mechanisms mediated by Pax5 during B has been shown to interact with the chromatin remodeling lymphopoiesis [19]. By using a ChIP-on-chip analysis, they complex Mi-2/NuRD in GC B cells, leading to the repression have identified Pax5 target genes in committed pro-B cells. of specific genes that are characteristic of plasma cells The authors also apply a proteomic approach to identify Pax5 [71, 72]. This Mi-2/NURD-mediated repression requires the interacting partners. They found that Pax5 interacts with the recruitment of histone deacetylases HDAC1 and HDAC2 members of the SWI/SNF chromatin remodeling complex [71, 72]. Brg1. BAF57 and BAF170. They also reported that PAX5 After activation of GC B cells Bcl-6 expression is down- recruits the NCoR1 repressor complex with its associated regulated in association with the expression of its target gene HDAC3 activity to repressed its target genes [19]. This study Blimp-1. Once expressed, Blimp-1 represses the gene expres- has provided novel important insight into the regulatory sion program of mature B cells, thereby promoting plasma network and epigenetic regulation, by which Pax5 directly cell differentiation [16, 17]. Mechanistically, Blimp-1 exerts controls B-cell commitment at the onset of B lymphopoiesis. its repressive transcriptional activity by recruiting regulators The mechanism by which Pax5 mediates transcriptional and coordinating epigenetic modifications at its target genes. repression of targets has also been informatively examined PRD1-BF1, the human orthologue of Blimp-1, silences using a candidate gene approach. One of the important target the interferon beta gene in response to viral infection by genes repressed by Pax5 in B cells is the colony-stimulating recruiting the histone methyltransferase (HMTase) G9a to factor receptor 1 gene (csf1r or c-fms), a gene essential for the interferon-beta promoter, resulting in H3K9me [73]. macrophage development. Csf1r is expressed at low levels in Blimp-1 has also been found in a complex with the arginine HSCs and downregulated in all nonmacrophage cell types. histone methyl transferase Prmt5, although the functional In HSCs, MPPs, CMPs, and CLPs the Csf1r promoter is significance of this interaction in B cells is not clear [74]. bound by transcription factors and its chromatin structure The histone lysine demethylase LSD1 has also been shown to in an active conformation [62]. However, the Csf1r gene interact with Blimp-1 [75]. Chromatin immunoprecipitation is silenced during B cell differentiation. Interestingly, an (ChIP) experiments indicated that Blimp-1 and LSD1 share intronic antisense transcription unit that is differentially some target genes leading to a more accessible chromatin regulated during lymphopoiesis overlaps with regions of structure [75]. Importantly, disruption of the Blimp-1-LSD1 de novo DNA methylation in B cells, highlighting DNA interaction resulted in attenuated antibody secretion of the

6 Comparative and Functional Genomics cells, highlighting the functional relevance of this interaction Itgam, Fcgr, and Csf1r for B cell function. In the last few years, additional transcription factors have H3K9/K14ac ↓H3K27me3 ↑ emerged as being involved in B cell terminal differentiation. H3K4me3 It has been reported that IRF4 and XBp1 control the mainte- myeloid-specific-genesmyeloid-specific-genes C/EBPC/EBPα nance of plasma cell identity. IRF4 is responsible for BLIMP- Pre-B No DNA methylationmethylation MacrophageMacrophage 1 induction and, in conjunction with XBp1, determines cell

changeschanges the fate of the plasma cell. The network of transcription ↑ H3K9/K14ac↑H3K27me3 factors Pax5, Bach2, and Bcl6 direct B cell development H3K4me3 into germinal center cells. It has been shown that Pax5 lymphoid-specific-geneslymphoid-specific-genes induces Bach2 expression after B cell activation, which in turn cooperates with Blc6 to repress Blimp-1 expression Vpreb1, Rag1, and Cd19 promoting activation-induced cytidine deaminase (AID) (a) Transdifferentiation expression and antibody class switch [76, 77]. MicroRNAs are also involved in the terminal differ- entiation of B cells. Peripheral B cells in transit to their ESCESC Pluripotency final maturation can give rise to two functionally distinct MatureMMatuurrree peripheral populations: follicular (FO) or marginal zone BBc ccelleelellllll Heterockaryon (MZ) B cells. FO versus MZ fate decision is functionally Polycomb repressor Eed coupled to BCR signaling and it has been suggested that B cells bearing BCRs with autoreactive specificities are Oct4, Sox2, preferentially driven into a MZ fate [78]. In 2010, Belver and Klf4 and c-Myc colleagues generated conditional Dicer-deficient mice at later Mature ↓PAX5 B cell iPS stages of B cell development [79]. They observed that miRNA ↑c/EBPα metabolism is important for such developmental stage since these mice presented an impairment in the generation of follicular B cells and an overrepresentation of marginal zone (b) Reprograming to pluripotency B cells. Accordingly, another phenotypic feature of these mice Figure 2: Transdifferentiation and reprogramming of B cells. was the presence of high titers of autoreactive antibodies (a) Ectopic expression of C/EBP in pre-B cells induces their [79]. They identified miR185 as an important factor for the transdifferentiation into macrophages. Epigenetic changes during correct BCR-mediated development of B cells. theprocessareshown.(b)Bcellscanbereprogrammedto pluripotency by fusion with ESCs (heterokaryon) or by transgenic induction of Oct4, Sox2, Klf4 and c-Myc (iPS). 6.BCellReprogrammingand Transdifferentiation Since 1987, when the possibility of reprogramming spe- the overexpression of C/EBPα [25](Figure 2). This cellu- cialized cells by the expression of a linage-specific tran- lar conversion has been considered a transdifferentiation scription factor was first reported, many studies have tried event since it is irreversible and does not require the to understand the molecular mechanisms that control all retrodifferentiation of pre-B cells to previous progenitor the processes involved. Due to the high developmental stages [81]. Using the cellular system generated in Graf’s complexity that characterizes the hematopoietic system, it laboratory, Radr´ıguez-Ubreva and colleagues performed a constitutes a model system with which study cell reprogram- high-throughput methylation analysis to study changes in ming and transdifferentiation in greater depth. In 1995, it DNA methylation during the transdifferentiation of pre-B was reported that overexpression of the erythroid lineage- cells into macrophages [82]. Surprisingly, they did not find specific transcription factor GATA-1 in myeloid leukemia any significant changes in DNA methylation during cellular cells induced their reprogramming into the megakary- conversion. However, they were able to identify the expected ocytic/erythroid lineage [24].Subsequently,Nuttetal. histone modifications in the genes that had previously been reported that Pax5-defective pro-B cells differentiated into described to be upregulated or downregulated during the functional macrophages, granulocytes, natural killer cells, process. In particular, they reported an increase in the osteoclasts, and dendritic cells when specific cytokines were enrichment of the active histone marks H3K9/K14ac and added to the culture medium [26]. Some years later, using H3K4me3, at the promoters of upregulated macrophage- knock-in and lineage-tracing technologies in mice, Xie specific genes, whereas a reduction of these modifications and colleagues were able to demonstrate in vivo repro- was observed in the B-cell-specific downregulated genes. In gramming of intrasplenic mature B cells into macrophages contrast, the repressive mark H3K27me3 was found to be by the overexpression of the myeloid transcription factor enriched in the B cell downregulated genes and reduced in C/EBPα [80]. More recently, the same laboratory generated the upregulated macrophage-specific genes [82](Figure 2). a robust reprogramming system in which murine pre- This study suggests that histone regulators are able to B cells were converted into functional macrophages by overcome the repressive effect of DNA methylation in

Comparative and Functional Genomics 7 macrophage-specific genes in the converted cells. It also with a draft of the “epigenetic-transcriptional” program that establishes an important difference from the process of controls B cell development and reprogramming. Finally, reprogramming towards pluripotency in which promoter conditional gene inactivation in mice will reveal the role DNA demethylation plays a crucial role. of specific epigenetic regulators during B cell development. In this regard, Hanna and colleagues demonstrated Thus, new regulatory networks connecting epigenetic and that pro-B and pre-B cells can be reprogrammed into transcription factors seem likely to be revealed in the context induced pluripotent stem (iPS) cells by the expression of of B lymphopoiesis in the future. the transcription factors Oct4, Sox2, Klf4, and c-Myc [27]. Interestingly, the expression of the four factors in mature Acknowledgments B cells does not result in the reprogramming of mature B cells to pluripotency. They found that expression of c/EBPα This work was supported by a grant from the Spanish in conjunction with the four “reprogramming” factors is Ministry of Science and Innovation (MICIIN) (BFU2008- necessary to generate iPS cells [27](Figure 2). iPS cell 00410) (to M.Parra) and a Marie Curie IRG Grant from lines derived from immature and mature B cells show the European Commission (FP7-MIRG-CT-2007-208119) promoter demethylation of the stem cell markers Oct4 and (to M.Parra). M. Parra was supported by a Ramon y Nanog, whereas both promoters are heavily methylated in Cajal contract from the Spanish Ministry of Science and the original B cells. Finally, in mature B cells the promoter Innovation (MICIIN). L.Roman-Gonzalez was supported by region of Pax5 shows high and low levels of enrichment an IDIBELL Ph.D. fellowship. T.Mahmoudi was supported for the active mark H3K4me3 and the repressive mark by an Erasmus M.C. Fellowship. H3K27me3, respectively. Conversely, equivalent enrichment of both histone modifications was observed in iPS lines References derived from mature B cells [27]. This study raises the challenging question of how B lymphocytes at different [1] J. Adolfsson, R. Mansson,˚ N. Buza-Vidas et al., “Identifi- developmental stages differ in their epigenetic landscape cation of Flt3+ lympho-myeloid stem cells lacking erythro- and how one factor can overcome this divergence to allow megakaryocytic potential: a revised road map for adult blood reprogramming into pluripotent cells. lineage commitment,” Cell, vol. 121, no. 2, pp. 295–306, 2005. A number of studies using experimental heterokaryons, [2] T. Yoshida, S. Yao-Ming Ng, J. C. Zuniga-Pflucker, and K. in which a somatic cell is reprogrammed towards pluripo- Georgopoulos, “Early hematopoietic lineage restrictions tency by fusion with mouse embryonic stem (ES) cells, directed by Ikaros,” Nature Immunology, vol. 7, no. 4, pp. 382– have also been used to reprogram B lymphocytes into 391, 2006. [3] H. Igarashi, S. C. Gregory, T. Yokota, N. Sakaguchi, and P. pluripotent cells. The laboratory of Amanda Fisher has W. Kincade, “Transcription from the RAG1 locus marks the shown that when mouse ES cells are fused with human B earliest lymphocyte progenitors in bone marrow,” Immunity, lymphocytes the expression of human pluripotent-associated vol. 17, no. 2, pp. 117–130, 2002. genes is rapidly induced [83]. Recently, the same group has [4] M. Kondo, I. L. Weissman, and K. Akashi, “Identification of elucidated some of the epigenetic mechanisms underlying clonogenic common lymphoid progenitors in mouse bone this reprogramming process. They showed that deletion of marrow,” Cell, vol. 91, no. 5, pp. 661–672, 1997. Eed, Suz12, Ezh2, and Ring1A/B, which are members of [5] C. Cobaleda and M. Busslinger, “Developmental plasticity of either the polycomb repressor complex PRC1 or PRC2, in lymphocytes,” Current Opinion in Immunology, vol. 20, no. 2, mouse ES cells abolishes their capacity to induce human pp. 139–148, 2008. B lymphocyte reprogramming towards pluripotency [84] [6] G. Natoli, “Maintaining cell identity through global control of (Figure 2). genomic organization,” Immunity, vol. 33, no. 1, pp. 12–24, 2010. [7] G. Bain, E. C. R. Maandag, D. J. Izon et al., “E2A proteins 7. Concluding Remarks are required for proper b cell development and initiation of immunoglobulin gene rearrangements,” Cell,vol.79,no.5,pp. The impressive advances in genome-wide methods and the 885–892, 1994. latest generation of ultrasequencing techniques are opening [8] G. Bain, E. C. Robanus Maandag, H. P. J. Te Riele et al., “Both up new, and challenging lines of research focused on the E12 and E47 allow commitment to the B cell lineage,” Immuni- elucidation of the epigenetic mechanisms underlying B cell ty, vol. 6, no. 2, pp. 145–154, 1997. differentiation and reprogramming. Many questions remain [9] Y. Zhuang, P. Soriano, and H. Weintraub, “The helix-loop- to be answered. Is there a specific “epigenetic signature” for helix gene E2A is required for B cell formation,” Cell, vol. 79, the different cellular states comprising B cell development? no. 5, pp. 875–884, 1994. ff How can lymphoid-specific transcription factors orchestrate [10] H. Lin and R. Grosschedl, “Failure of B-cell di erentiation in the epigenetic machinery at different genes and genome mice lacking the transcription factor EBF,” Nature, vol. 376, ff no. 6537, pp. 263–267, 1995. regions to facilitate the choice to di erentiate into a particu- [11] H. S. Dengler, G. V. Baracho, S. A. Omori et al., “Distinct func- lar cellular lineage? Is the expression of epigenetic regulators tionss for the transcription factor Foxo1 at various stages of lineage-specific? Epigenetic modification analyses, genome- Bcelldifferentiation,” Nature Immunology, vol. 9, no. 12, pp. wide RNA and ChIP-Seq studies, quantitative proteomics, 1388–1398, 2008. and systematic functional studies offer us the opportunity [12] P. Urbanek, Z. Q. Wang, I. Fetka, E. F. Wagner, and M. to obtain high-quality measurements that will provide us Busslinger, “Complete block of early B cell differentiation and

8 Comparative and Functional Genomics

altered patterning of the posterior midbrain in mice lacking [30] T. Kouzarides, “Chromatin modifications and their function,” Pax5/BSAP,” Cell, vol. 79, no. 5, pp. 901–912, 1994. Cell, vol. 128, no. 4, pp. 693–705, 2007. [13] D. Allman, A. Jain, A. Dent et al., “BCL-6 expression during [31] J. K. Wiencke, S. Zheng, Z. Morrison, and R. F. Yeh, “Dif- B-cell activation,” Blood, vol. 87, no. 12, pp. 5257–5268, 1996. ferentially expressed genes are marked by histone 3 lysine 9 [14] A. L. Dent, A. L. Shaffer, X. Yu, D. Allman, and L. M. Staudt, trimethylation in human cancer cells,” Oncogene, vol. 27, no. “Control of inflammation, cytokine expression, and germinal 17, pp. 2412–2421, 2008. center formation by BCL-6,” Science, vol. 276, no. 5312, pp. [32] C. R. Vakoc, S. A. Mandat, B. A. Olenchock, and G. A. Blobel, 589–592, 1997. “Histone H3 lysine 9 methylation and HP1γ are associated [15] B. H. Ye, G. Cattoretti, Q. Shen et al., “The BCL-6 proto- with transcription elongation through mammalian chroma- oncogene controls germinal-centre formation and Th2- type tin,” Molecular Cell, vol. 19, no. 3, pp. 381–391, 2005. inflammation,” Nature Genetics, vol. 16, no. 2, pp. 161–170, [33] K. Georgopoulos, M. Bigby, J. H. Wang et al., “The Ikaros gene 1997. is required for the development of all lymphoid lineages,” Cell, [16] A. L. Shaffer, K. I. Lin, T. C. Kuo et al., “Blimp-1 orchestrates vol. 79, no. 1, pp. 143–156, 1994. plasma cell differentiation by extinguishing the mature B cell [34] J. H. Wang, A. Nichogiannopoulou, L. Wu et al., “Selective gene expression program,” Immunity, vol. 17, no. 1, pp. 51–62, defects in the development of the fetal and adult lymphoid 2002. system in mice with an Ikaros null mutation,” Immunity, vol. [17] R. Sciammas and M. M. Davis, “Modular nature of blimp-1 5, no. 6, pp. 537–549, 1996. in the regulation of gene expression during B cell maturation,” [35] S. Y. M. Ng, T. Yoshida, and K. Georgopoulos, “Ikaros and Journal of Immunology, vol. 172, no. 9, pp. 5427–5440, 2004. chromatin regulation in early hematopoiesis,” Current Opin- [18] Y. C. Lin, S. Jhunjhunwala, C. Benner et al., “A global network ion in Immunology, vol. 19, no. 2, pp. 116–122, 2007. of transcription factors, involving E2A, EBF1 and Foxo1, that [36] J. Kim, S. Sif, B. Jones et al., “Ikaros DNA-binding proteins orchestrates B cell fate,” Nature Immunology, vol. 11, no. 7, pp. direct formation of chromatin remodeling complexes in lym- 635–643, 2010. phocytes,” Immunity, vol. 10, no. 3, pp. 345–355, 1999. [19] S. McManus, A. Ebert, G. Salvagiotto et al., “The transcription [37] J. Zhang, A. F. Jackson, T. Naito et al., “Harnessing of the factor Pax5 regulates its target genes by recruiting chromatin- nucleosome-remodeling-deacetylase complex controls lym- modifying proteins in committed B cells,” EMBO Journal, vol. phocyte development and prevents leukemogenesis,” Nature 30, no. 12, pp. 2388–2404, 2011. Immunology, vol. 13, no. 1, pp. 86–94, 2012. [20] N. Novershtern, A. Subramanian, L. N. Lawton et al., “Densely [38] S. R. McKercher, B. E. Torbett, K. L. Anderson et al., “Targeted interconnected transcriptional circuits control cell states in disruption of the PU.1 gene results in multiple hematopoietic human hematopoiesis,” Cell, vol. 144, no. 2, pp. 296–309, abnormalities,” EMBO Journal, vol. 15, no. 20, pp. 5647–5658, 2011. 1996. [21] H. Gao, K. Lukin, J. Ram´ırez, S. Fields, D. Lopez, and J. [39] E. W. Scott, M. C. Simon, J. Anastasi, and H. Singh, “Require- Hagman, “Opposing effects of SWI/SNF and Mi-2/NuRD ment of transcription factor PU.1 in the development of mul- chromatin remodeling complexes on epigenetic reprogram- tiple hematopoietic lineages,” Science, vol. 265, no. 5178, pp. ming by EBF and Pax5,” Proceedings of the National Academy 1573–1577, 1994. of Sciences of the United States of America, vol. 106, no. 27, pp. [40] S. Heinz, C. Benner, N. Spann et al., “Simple combinations 11258–11263, 2009. of lineage-determining transcription factors prime cis-regula- [22] K. Walter, C. Bonifer, and H. Tagoh, “Stem cell-specific epige- tory elements required for macrophage and B cell identities,” netic priming and B cell-specific transcriptional activation at Molecular Cell, vol. 38, no. 4, pp. 576–589, 2010. the mouse Cd19 locus,” Blood, vol. 112, no. 5, pp. 1673–1682, [41] M. Ye and T. Graf, “Early decisions in lymphoid development,” 2008. Current Opinion in Immunology, vol. 19, no. 2, pp. 123–128, [23] I. H. Su and A. Tarakhovsky, “Epigenetic control of B cell dif- 2007. ferentiation,” Seminars in Immunology, vol. 17, no. 2, pp. 167– [42] M. Busslinger, “Transcriptional control of early B cell develop- 172, 2005. ment,” Annual Review of Immunology, vol. 22, pp. 55–79, 2004. [24]H.Kulessa,J.Frampton,andT.Graf,“GATA-1reprograms [43] A. S. Lazorchak, J. Wojciechowski, M. Dai, and Y. Zhuang, avian myelomonocytic cell lines into eosinophils, throm- “E2A promotes the survival of precursor and mature B lym- boblasts, and erythroblasts,” Genes and Development, vol. 9, phocytes,” Journal of Immunology, vol. 177, no. 4, pp. 2495– no. 10, pp. 1250–1262, 1995. 2504, 2006. [25] L. H. Bussmann, A. Schubert, T. P. Vu Manh et al., “A robust [44] R. H. Amin and M. S. Schlissel, “Foxo1 directly regulates the and highly efficient immune cell reprogramming system,” Cell transcription of recombination-activating genes during B cell Stem Cell, vol. 5, no. 5, pp. 554–566, 2009. development,” Nature Immunology, vol. 9, no. 6, pp. 613–622, [26] S. L. Nutt, B. Heavey, A. G. Rolink, and M. Busslinger, 2008. “Commitment to the B-lymphoid lineage depends on the [45] H. Maier, R. Ostraat, H. Gao et al., “Early B cell factor coop- transcription factor Pax5,” Nature, vol. 401, no. 6753, pp. 556– erates with Runx1 and mediates epigenetic changes associated 562, 1999. with mb-1 transcription,” Nature Immunology, vol. 5, no. 10, [27] J. Hanna, S. Markoulaki, P. Schorderet et al., “Direct repro- pp. 1069–1077, 2004. gramming of terminally differentiated mature B lymphocytes [46] E. Mercer, Y. Lin, C. Benner et al., “Multilineage priming of to pluripotency,” Cell, vol. 133, no. 2, pp. 250–264, 2008. enhancer repertoires precedes commitment to the B and mye- [28] C. R. Clapier and B. R. Cairns, “The biology of chromatin loid cell lineages in hematopoietic progenitors,” Immunity, vol. remodeling complexes,” Annual Review of Biochemistry, vol. 35, no. 3, pp. 413–425, 2011. 78, pp. 273–304, 2009. [47] T. Treiber, E. M. Mandel, S. Pott et al., “Early B cell factor 1 [29] S. Guil and M. Esteller, “DNA methylomes, histone codes regulates B cell gene networks by activation, repression, and and miRNAs: tying it all together,” International Journal of transcription- independent poising of chromatin,” Immunity, Biochemistry and Cell Biology, vol. 41, no. 1, pp. 87–95, 2009. vol. 32, no. 5, pp. 714–725, 2010.

Comparative and Functional Genomics 9

[48] X.-X. Jiang, Q. Nguyen, Y. Chou et al., “Control of B cell devel- [65] K. Rajewsky, “Clonal selection and learning in the antibody opment by the histone H2A deubiquitinase MYSM1,” Immu- system,” Nature, vol. 381, no. 6585, pp. 751–758, 1996. nity, vol. 35, no. 6, pp. 883–896, 2011. [66] M. Kraus, M. B. Alimzhanov, N. Rajewsky, and K. Rajewsky, [49] D. O’Carroll, I. Mecklenbrauker, P. P. Das et al., “A Slicer- “Survival of resting mature B lymphocytes depends on BCR independent role for Argonaute 2 in hematopoiesis and the signaling via the Igα/β heterodimer,” Cell, vol. 117, no. 6, pp. microRNA pathway,” Genes and Development, vol. 21, no. 16, 787–800, 2004. pp. 1999–2004, 2007. [67] K. L. Calame, K. I. Lin, and C. Tunyaplin, “Regulatory mecha- [50] S. B. Koralov, S. A. Muljo, G. R. Galler et al., “Dicer ablation nisms that determine the development and function of plasma affects antibody diversity and cell survival in the B lymphocyte cells,” Annual Review of Immunology, vol. 21, pp. 205–230, lineage,” Cell, vol. 132, no. 5, pp. 860–874, 2008. 2003. [51] C. Z. Chen, L. Li, H. F. Lodish, and D. P. Bartel, “MicroRNAs [68] T. Honjo, M. Muramatsu, and S. Fagarasan, “Aid: how does it modulate hematopoietic lineage differentiation,” Science, vol. aid antibody diversity?” Immunity, vol. 20, no. 6, pp. 659–668, 303, no. 5654, pp. 83–86, 2004. 2004. [52]B.Zhou,S.Wang,C.Mayr,D.P.Bartel,andH.F.Lodish, [69]R.Reljic,S.D.Wagner,L.J.Peakman,andD.T.Fearon,“Sup- “miR-150, a microRNA expressed in mature B and T cells, pression of signal transducer and activator of transcription 3- blocks early B cell development when expressed prematurely,” dependent B lymphocyte terminal differentiation by BCL-6,” Proceedings of the National Academy of Sciences of the United Journal of Experimental Medicine, vol. 192, no. 12, pp. 1841– States of America, vol. 104, no. 17, pp. 7080–7085, 2007. 1847, 2000. [53] C. Xiao, D. P. Calado, G. Galler et al., “MiR-150 controls B cell [70] A. L. Shaffer,X.Yu,Y.He,J.Boldrick,E.P.Chan,andL.M. differentiation by targeting the transcription factor c-Myb,” Staudt, “BCL-6 represses genes that function in lymphocyte Cell, vol. 131, no. 1, pp. 146–159, 2007. differentiation, inflammation, and cell cycle control,” Immu- [54] A. Ventura, A. G. Young, M. M. Winslow et al., “Targeted nity, vol. 13, no. 2, pp. 199–212, 2000. deletion reveals essential and overlapping functions of the [71]N.Fujita,D.L.Jaye,M.Kajita,C.Geigerman,C.S.Moreno, miR-17∼92 family of miRNA clusters,” Cell, vol. 132, no. 5, and P. A. Wade, “MTA3, a Mi-2/NuRD complex subunit, pp. 875–886, 2008. regulates an invasive growth pathway in breast cancer,” Cell, [55] R. M. O’Connell, D. S. Rao, A. A. Chaudhuri, and D. Balti- vol. 113, no. 2, pp. 207–219, 2003. more, “Physiological and pathological roles for microRNAs in [72]N.Fujita,D.L.Jaye,C.Geigermanetal.,“MTA3andtheMi- the immune system,” Nature Reviews Immunology, vol. 10, no. 2/NuRD complex regulate cell fate during B lymphocyte dif- 2, pp. 111–122, 2010. ferentiation,” Cell, vol. 119, no. 1, pp. 75–86, 2004. [56] S. Kuchen, W. Resch, A. Yamane et al., “Regulation of MicroR- [73]I.Gyory,J.Wu,G.Fejer,E.Seto,andK.L.Wright,“PRDI-BF1´ NA expression and abundance during lymphopoiesis,” Immu- recruits the histone H3 methyltransferase G9a in transcrip- nity, vol. 32, no. 6, pp. 828–839, 2010. tional silencing,” Nature Immunology, vol. 5, no. 3, pp. 299– [57] C. Cobaleda, A. Schebesta, A. Delogu, and M. Busslinger, 308, 2004. “Pax5: the guardian of B cell identity and function,” Nature [74] K. Ancelin, U. C. Lange, P. Hajkova et al., “Blimp1 associates Immunology, vol. 8, no. 5, pp. 463–470, 2007. with Prmt5 and directs histone arginine methylation in mouse [58] T. Decker, M. Pasca di Magliano, S. McManus et al., “Stepwise germ cells,” Nature Cell Biology, vol. 8, no. 6, pp. 623–630, activation of enhancer and promoter regions of the B cell 2006. commitment gene Pax5 in early lymphopoiesis,” Immunity, [75] S. T. Su, H. Y. Ying, Y. K. Chiu, F. R. Lin, M. Y. Chen, and K. vol. 30, no. 4, pp. 508–520, 2009. I. Lin, “Involvement of histone demethylase LSD1 in blimp-1- [59] A. Delogu, A. Schebesta, Q. Sun, K. Aschenbrenner, T. Perlot, mediated gene repression during plasma cell differentiation,” and M. Busslinger, “Gene repression by Pax5 in B cells is Molecular and Cellular Biology, vol. 29, no. 6, pp. 1421–1431, essential for blood cell homeostasis and is reversed in plasma 2009. cells,” Immunity, vol. 24, no. 3, pp. 269–281, 2006. [76] A. Muto, K. Ochiai, Y. Kimura et al., “Bach2 represses plasma [60] A. Schebesta, S. McManus, G. Salvagiotto, A. Delogu, G. A. cell gene regulatory network in B cells to promote antibody Busslinger, and M. Busslinger, “Transcription factor Pax5 acti- class switch,” EMBO Journal, vol. 29, no. 23, pp. 4048–4061, vates the chromatin of key genes involved in B cell signaling, 2010. adhesion, migration, and immune function,” Immunity, vol. [77] S. L. Nutt, N. Taubenheim, J. Hasbold, L. M. Corcoran, and 27, no. 1, pp. 49–63, 2007. P. D. Hodgkin, “The genetic network controlling plasma cell [61] C. Pridans, M. L. Holmes, M. Polli et al., “Identification of differentiation,” Seminars in Immunology,vol.23,no.5,pp. Pax5 target genes in early B cell differentiation,” Journal of 341–349, 2011. Immunology, vol. 180, no. 3, pp. 1719–1728, 2008. [78] D. Allman and S. Pillai, “Peripheral B cell subsets,” Current [62] H. Tagoh, A. Schebesta, P. Lefevre et al., “Epigenetic silencing Opinion in Immunology, vol. 20, no. 2, pp. 149–157, 2008. of the c-fms locus during B-lymphopoiesis occurs in discrete [79]L.Belver,V.G.deYebenes,´ and A. R. Ramiro, “MicroRNAs steps and is reversible,” EMBO Journal, vol. 23, no. 21, pp. prevent the generation of autoreactive antibodies,” Immunity, 4275–4285, 2004. vol. 33, no. 5, pp. 713–722, 2010. [63] H. Tagoh, R. Ingram, N. Wilson et al., “The mechanism of [80]H.Xie,M.Ye,R.Feng,andT.Graf,“Stepwisereprogramming repression of the myeloid-specific c-fms gene by Pax5 during of B cells into macrophages,” Cell, vol. 117, no. 5, pp. 663–676, B lineage restriction,” EMBO Journal, vol. 25, no. 5, pp. 1070– 2004. 1080, 2006. [81]A.DiTullio,T.P.VuManh,A.Schubert,R.Mansson,˚ and T. [64] V. Giambra, S. Volpi, A. V. Emelyanov et al., “Pax5 and linker Graf, “CCAAT/enhancer binding protein α (C/EBPα)-induced histone H1 coordinate DNA methylation and histone modi- transdifferentiation of pre-B cells into macrophages involves fications in the 3 regulatory region of the immunoglobulin no overt retrodifferentiation,” Proceedings of the National heavy chain locus,” Molecular and Cellular Biology, vol. 28, no. Academy of Sciences of the United States of America, vol. 108, 19, pp. 6123–6133, 2008. no. 41, pp. 17016–17021, 2011.

10 Comparative and Functional Genomics

[82] J. Rodr´ıguez-Ubreva, L. Ciudad, D. Gomez-Cabrero´ et al., “Pre-B cell to macrophage transdifferentiation without sig- nificant promoter DNA methylation changes,” Nucleic Acids Research, vol. 40, no. 5, pp. 1954–1968, 2012. [83] C. F. Pereira, R. Terranova, N. K. Ryan et al., “Heterokaryon- based reprogramming of human B lymphocytes for pluripo- tency requires Oct4 but not Sox2,” PLoS Genetics, vol. 4, no. 9, Article ID e1000170, 2008. [84] C. F. Pereira, F. M. Piccolo, T. Tsubouchi et al., “ESCs require PRC2 to direct the successful reprogramming of differentiated cells toward pluripotency,” Cell Stem Cell, vol. 6, no. 6, pp. 547– 556, 2010.