Nuclear regulatory landscape of the circadian clock in mouse liver

Jingkui Wang1,, Daniel Mauvoisin2, Eva Martin2, Florian Atger2, Antonio Nunes Galindo2, Federico Sizzano2, Loïc Dayon2, Martin Kussmann2, Patrice Waridel3, Manfredo Quadroni3, Felix Naef1 and Frédéric Gachon2

1Institute of Bioengineering, School of Life Sciences, Ecole Polytechnique Fédérale de Lausanne

and Swiss Institute of Bioinformatics, Lausanne, Switzerland 2Nestlé Institute of Health Sciences, Lausanne, Switzerland.

3Protein Analysis Facility, University of Lausanne, Lausanne, Switzerland.

Abstract Diurnal oscillations of gene expression dictated by the circadian clock enable living organisms to coordinate their physiological processes with daily environmental changes. Although such rhythms have been extensively studied at the level of transcription and mRNA accumulation, comparably little is known at the proteins level, though recent proteomics studies indicated that total protein rhythms generally appeared damped compared to their cognate mRNAs. In order to further dissect how diurnal rhythms affect key functions such as transcription or chromatin remodeling, we quantified the temporal nuclear accumulation of proteins and phosphoproteins from mouse liver by SILAC-based MS. Our analysis identified ~5000 nuclear proteins, including all core-clock and clock-related proteins, over 500 of which are found to be rhythmic under a stringent statistical threshold (FDR <5%). These rhythmic nuclear proteins are mainly controlled at the post-transcriptional level and are often parts of complexes showing robust diurnal nuclear accumulation. These rhythmic complexes are notably involved in transcriptional regulation, rRNA synthesis, ribosome assembly, as well as DNA damage repair. From the parallel analysis of the nuclear phospho-proteome, we could infer the temporal activity o f k i n a s e s c o n t r i b u t i n g t o t h e s e r h y t h m i c phosphorylations. A large fraction of the kinase activities were implicated in cell signaling and cell cycle regulation. In addition, 80 transcription factors and about 100 transcriptional coregulators showed clear diurnal oscillations in the nucleus, enlarging the extent of transcriptional and epigenetic regulations by the circadian clock and/or systemic cues. Finally, a number of proteins with functions in the cytoplasm are detected in the nucleus at a common and sharp time near the night-day transition. This phenomenon is probably linked to the rhythmic endoreplication occurring in hepatic cells and associated to leakage of the nuclear membrane. Taken together, these findings provide unprecedented insights into the regulatory landscape of the diurnal liver nucleus.

0 12 24 36 45

Every 3 hours 2 WT mice

0 12 24

Every 12 hours 8 SILAC mice

SILAC mix (1:1:1:1:1:1:1:1:1:1:1:1:1:1:1:1) WT mix (1:1)

mix (1:1)

Trypsin Digest Fractionation

Unlabeled mice

F2

13C6-Lysine

MS analysis

0 3 6 9 12 15 18 21 24 27 30 33 36 39 42 45

Methods & Results

Nuclei purification and proteins extraction

Phospho-enrichment

MS analysis

1)  5-10% of total proteins show diurnal accumulations. 2)  ~50% of these rhythmic proteins do not have corresponding

rhythmic mRNAs and are highly enriched in secretory proteins. Question: How about the rhythmicity of proteins in different subcompartments such as nucleus ?

CoveragePercent.rhythmic

0.0

0.2

0.4

0.6

0.8

1.0

Fig. 1 ~ 5000 nuclear proteins were quantified by SILAC-based MS including

core-clock and clock-related proteins

TF

Cor

egul

ator

RN

A pr

oces

sing

Kin

ase

Pho

spha

tase

4820 proteins quantified

Nucleus Both Cytoplasm

mRNA protein mRNA protein mRNA protein

rhyt

hmic

no

n-rh

ythm

ic

FASN  

ALB  

ACACA  

LC  

NE  

LC  

NE  

LC  

NE  

αβGSK3  

LC  

NE  SIRT7  LC  

NE  

NR3C1  LC  

NE  

Fig. 3 Subunits of nuclear protein complexes showed highly similar and

diurnal accumulations

Fig. 5 Transcription factors and coregulators (FDR<5%) involved in the

circadian transcription

FASN ZT0 confocal

Lamin A/C IF

P-CDC2

LC

NE

T-MCM2

LC

NE

P-­‐CDC2  

MCM2  

LC  

NE  

LC  

NE  

3 6 9 12 15 18 21 24 27 30 33 36 39 42 45 ZT 0

A) Distribution of samples numbers quantified for detected nuclear proteins. B) Coverage of nuclear proteins and percentages of rhythmic ones

(FDR<5%) for different functional categories. C) Phases and peak-trough amplitudes of core-clock and clock-related

proteins with examples in D).

A B

C

A) Phase distribution for the rhythmic nuclear proteins (FDR<0.05) grouped by their annotated localizations (UNIPROT) B) Heat maps of the rhythmic proteins and their corresponding mRNAs. Data is standardized by rows and gray blocks indicate missing data. C) Western blot of individual rhythmic proteins performed on nuclear extract. The graphs represents the quantification of the blots and the corresponding mass spec data.

A

B

C

Fig. 2 Rhythmic nuclear proteins are mainly post-transcriptionally regulated

Take-home messages

Fig. 6 Rhythmic endoreplication probably causes leakage of cytoplasmic proteins by weakening nuclear envelope

A

B

A

B

C LC  

NE  CDC6  

P27  Kip1  LC

NE

LC  

NE  

A) FACS analyses of nuclei isolated from mouse liver around the clock and separated into ploidy populations by (n = 4 light dark cycles) B) Localization of FASN at ZT0 (confocal fluorescence) and immunofluorescence of Lamin A/C in purified liver nuclei. C) Western blot of individual protein involved in cell cycle regulation and DNA replication.

A) Peak phases of rhythmic nuclear protein complexes and individual examples are found in B).

1. SILAC nuclear proteomics in mouse liver -  High resolution (~5000 proteins), -  ~12% highly rhythmic (FDR<5%). -  Almost all core clock and clock

related genes identified and quantified.

2. Within these rhythmic proteins -  81 transcription factors and more than

100 coregulators : most of them are new.

-  Many nuclear protein complexes also display diurnal expression.

3. Annotated cytoplasmic proteins are also

-  detected in the nucleus with a sharp day night transition phase.

è may be due to a weakening of the nuclear envelope resulting from rhythmic endoreplication/replication.

Previous work

SILAC mass spectrometry analysis of mouse liver nuclei

nb of quantified samples

nb o

f pro

tein

s

0 5 10 150

500

1000

1500

2000

2500

3000

●NFIL3

●ARNTL,CLOCK

●

●NR1D1

●NR1D2

●DBP

●HLF

●TEF

●PER1●PER2

●CRY2

●RORC

● RORA

●CRY1

●

●

●

core clockstablizing loopoutput regulators

ZT0

ZT6

ZT12

ZT18

D ARNTL

●

●●

●

●

●

●●

●●

●●

0 6 12 18 24

−0.6

−0.3

0.0

0.3

CLOCK

●

●

●

●

●

● ●

●

●●

● ●

0 6 12 18 24

−0.5

−0.2

0.1

0.4

PER1

●

●

●●

●

●

● ●

●

●

●

●

0 6 12 18 24−2

−1

0

1

PER2

●

●

●

●

●

● ●

●

0 6 12 18 24−2

−1

0

1

Fig.4 Rhythmic activities of kinases predicted by nuclear phospho-proteome

nucleus    phospho  C    A B

C

A) and B) Examples and heat maps of rhythmic nuclear phospho-proteins with non-rhythmic nuclear proteins. C) Peak phases of rhythmic kinase activities predicted by the nuclear phospho-proteins, some of which are confirmed by their nuclear protein accumulations in D)

ZT6  ZT18  

ZT0  

ZT12  

NFIA, S280

●

●

●

●

●

●

●

● ● ●

●●

●

● ●●

●

●

●

●●

●

● ●

●

●

●

●●

●

●

●

0 12 24 36 48−0.7

−0.4

−0.1

0.2

0.5

0.8CIC, S1080

●

●

●

●

● ●

●●

●●

●

●

●

● ●

●●

●

●●

●

●

●

●

●

●

●

●

●

●● ●

0 12 24 36 48−0.9

−0.6

−0.3

0.0

0.3

0.6

NCBP1, S22

● ●●

● ● ●●

●● ●

● ●

●

● ●●

●

● ●

●

●

●

●

●

●

●

●

●

● ●●

●

0 12 24 36 48−0.7

−0.4

−0.1

0.2

0.5

0.8

CHD4, S1517

●● ●

●● ●

● ●● ● ●

●

●●

●●

●

●

●

●

●

●

●

●

●

●●

● ●

●

●●

0 12 24 36 48

−0.5−0.3−0.10.10.30.50.7

D ●

●

● ●

●

●

● ●●

●

●●

●

●

●

●

CSNK1D

0 12 24 36 48

−1.0−0.40.20.81.4

●

●

●●

●

● ●

●

●

●

●

●

●

●

●

●

GSK3A

0 12 24 36 48

−2.0−1.2−0.40.41.2

● ●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

MAPK14

0 12 24 36 48

−1.2−0.40.41.2

● ●

●

●

●

●

●

●

●

●●

●

●

●

●

●

PRKAA2

0 12 24 36 48

−1.0−0.40.20.81.42.0

REP

IN1

DIDO

1NF

IL3

AATF

CIZ1

MEF

2DHNF1

BPHC2

PPARD

ILF2NR2F6

ZKSCAN3

IRF9HNF1A

TBX3ZMYND8GTF2IRD1CLOCKARNTLTFEB

ELF1CICATF1IKZF1NFIATCF7L2FLI1JUNBZFHX3FOXP1RFX1HBP1ELF2NR1D1

NR2C1

ZFHX4

MGAE4F1

SP2NFIBHM

BOX1

SCMH1

CCNT2

FOXP2

USF2TERF1N

R1D

2ZN

F187PPAR

AD

BPAR

ID5BR

XRA

PATZ

1H

LFN

R3C

2KLF1

3TE

F

NR3C

1

BHLH

E40

MAFB

ERF

FOXA3

NR1H4NFIX

HCFC1PCBP1

HSF1FOXK1FOXA1

KLF3

SREBF1

ESRRAETV3

HNF4A

RORC

TCF12

RORA

CEBPGESR1

TP53

●

IRF2

BP1

●

ZFPM

1

●

CAC

YBP

●

PAF1

●

ZFP1

28

●

CXXC

5

●

SETD

8

●

BCOR

●

EDF1

●

NRIP1

●

ZBTB

33

●

MAML1

●

IRF2

BP2

●

RB1

●DDX17

● WIZ

● CBX2

● ZMYND8

● CBFA2T2

● TBL1X

● PHF21A● EHMT2● SETDB1

● KAT5● JMJD1C●ATF7IP

●MTA2●NCOR1●TBL1XR1●HDAC3

●CABIN1●RBBP7●EMSY

●N4BP2L2●MLLT10

●ZBTB1

●BPTF

●CHD6

●EP400

●KDM3A

●HIRA

●MED12

●MTA3

●NACC1

●MRGBP

●KDM5A

●DAXX

●TINF2

●MED28

●EPC2

●TRRAP

●PPARGC1B

●SIN3A

●MED

23

●CIPC

●SAP130

●CR

EBBP

●CEC

R2

●BAZ2A

●MED

13

●

WBP

7

●

PHF1

7

●

MED

16

●

ASH

1L

●

EP30

0

●

MLL

3

●

VPS7

2

●

MED

20

●

DDX5

●

LIN5

4

●

PSIP

1

●

ARID

2

●

BAZ2

B

●

LIN3

7

●

THRA

P3

●

BAP1

●

ATRX

●

NIPB

L

●

RAD5

4L2

●

TAF2

●

TRIM

24

●

AFF4

●

CRTC2

●PER3

●PER1 ●PER2

●

CRY2 ●

NONO

●

BRD2

●

BAZ1B

●

ABT1 ●

CRY1 ●

YAF2

●

BCCIP ●

SKI ●

KDM

1B

●

SUPT6H

●

HDAC

5

TF.nuclearTF.bothTF.phospho

●●●●●●●

CoregulatorAcetylDeacetylMethylDemethylUbiqDeubiq

M NMA

Arp2/3Arpc2Actr2Arpc1bActr3Arpc3Arpc4Arpc5

0 12 24 36 48−1.8

−0.8

0.2

1.2

2.2CSN/CSA/DDB2

Polr2aCops3Cops5Cops8Cops4Cops2Gps1Cops6Cops7aDdb1Rbx1Cops7bGferCul4aDdb2

0 12 24 36 48−3

−2

−1

0

1

2

NuRDRbbp7Mta2Mta3Zfpm1Gatad2bChd4Smarca4Mbd3Hdac2Mta1Smarcb1Arid1aMbd2Smarcd2Hdac1;Gm10093Dpf2Actl6aSmarcc2Smarcc1Mbd3Rbbp4HnrnpcSmarce1

0 12 24 36 48−2.8

−1.8

−0.8

0.2

1.2

2.2

TRAP/SMCC/DRIPMed16Thrap3Med13TrrapMed12Esr1Med23Cdk8CcncMed21Med17Med1Med24RxrbCcnt1Med14Med10Med6Kat2aMed27Med26Med4Brd4Med7Cdk9Med31

0 12 24 36 48−3.5

−2.5

−1.5

−0.5

0.5

1.5

2.5Mediator/asscociated

Med16Thrap3Med13Med12Med28Med20;Gm20517Med23Cdk8CcncMed21Med17Med1Med24Med25Med29Med13lMed15Med14Med10Med6Acad8Med27Med26Med22Med4Brd4Opa1Med19Med11Med18Med8Med7Cdk19Med31

0 12 24 36 48−3.5−2.5−1.5−0.50.51.52.5

Rev−ErbA−Ncor1−Hdac3Nr1d1Ncor1Hdac3Ncor1

0 12 24 36 48−2.1

−1.1

−0.1

0.9

1.9NCOR/SMRT

Ncor1Tbl1xr1Tbl1xHdac3Zbtb33Ncor2Ncor2Trim33Ncor1Gps2

0 12 24 36 48−2.3

−1.3

−0.3

0.7

1.7

2.7PER

Cry2Per1Cry1Per2Per3NonoWdr5

0 8 16 24 32 40 48

−2.1

−1.1

−0.1

0.9

1.9

●

AP2

●

Arp2/3

●

Profilin 1

●

Endo

cytic

coa

t

●

RICH

1/AM

OT p

olar

ity

●

20S

met

hyltr

ansfe

rase

●

KCNQ

1

●

CLIC

4

●

AQP2−f

orce−g

ener

ator

●

Neph

rin

●

WIP−W

ASp−

actin−m

yosin−I

ia

●

Emer

in

● DNMT1−RB1−HDAC1−E2F1

● IMP3−IMP4−MPP

● Exosome

● RNA PolI

● Bmal1−Clock

●CoREST−HDAC

●NCOR/SMRT●TFC4−CTNNB1●SETDB1−HMTase●HDAC1/2−associated●DNA ligase III−XRCC1

●NuRD●Ikaros●Rev−ErbA−Ncor1−Hdac3

●NCOR−SIN3

●HIRA●NURF●SIN3−associated

●APC/C

●NuA4/Tip60 HAT

●ZNF198−PML

●MRE11−RAD50−NBN

●TRF1

●p300−CBP

●SWI/SNF

●TRAP/SMCC/DRIP

●CDK8−CCNC−RNA PolII

●Mediator/asscociated

●

Bmal1−C

lock−

Pers

●

PA28

●

PER

●CPSF ●

PSF−P54 ●

Parvulin pre−rRNP ●

WICH ●CKII ●

p130Cas−ER−alpha−cSrc ●

RNA PolII/associated ●

Core Cohensin ●

Retromer ●

HSP90−associated ●

AMPK

●

Alpha−GDI−Hsp9●

MNK1−eIF4F●

Mt3−Hsp84−Ck●

AP1●

ESCRT−II●

APPBP1−UBA3●

SURF●

CSN/CSA/DDB2●

CCT

●

●

●

●

●

●

●

●

transcriptioncytoskeletonchaperonecell cycleDNA repairproteolisisprotein transprotothers

PER

ZT6

ZT12

ZT18

ZT0

B ●E−Box

●

RRE

ClockArntl

Nr1d1

Nr1d2

Rorc

Rora

ActivatorRepressor

●D−Box

●KLFE

●E−Box

Nfil3

DbpH

lf

Klf13Tef

Bhlhe40

Klf3 ●HNF4A_NR2Fs

●IR3

Nr2f6

Nr3c2Nr3c1

Esrra

Esr1

●

HNF1s● TFE

B.m

● CRE● FOXE

●RXRE

●

NR1H4.

m●

HSE

●HNF4A_NR2Fs

●E−Box

Hnf1b

Hnf1a

Tfeb

Atf1Foxp1

Foxp2

Rxra

Foxa3

Nr1h4

Hsf1Foxk1Foxa1

Hnf4a

Tcf12

●

ETS

Elf1Fli1

Elf2Erf

Etv3

core-clock output-regulator hormone cues Ets family metabolic cues

0 1 2345678

91011121314

1516171819202122 23

0 1 2345678

91011121314

1516171819202122 23 0 1 2

345678

91011121314

1516171819202122 23

●

Map4k2_ST

●

Mapk14_ST

●

Mapk15_ST

●

Plk2_ST

●

Mapk9_ST

●Stk16_

ST

● Gsk3a_

ST

● Mapk1_ST● Sik1_ST● Dstyk_ST●Ripk3_ST●Camkk2_ST●Nek9_ST●Tesk2_ST●Cdk5_ST●Mapk3_ST●Tgfbr2_ST

●Cdk6_ST●Mapk12_ST●Mapkapk5_ST

●Acvr2b_ST●Prkcb_ST●Gsk3b_ST

●Mapk7_ST

●Prkx_ST●Raf1_ST

●Wee1_ST

●Mapk8_ST

●Nuak1_ST

●Map3k13_ST

●Rps6ka3_ST

●Cdk4_ST

●

Uhmk1_ST

●

Camk2d_ST

●

Cdk7_ST

●Clk1_

ST●

Stk24_ST

●Map3k8_ST●Dyrk1b_ST●

Csnk2a2_ST ●Akt1_ST ●Cdk16_ST ●Srpk2_ST ●Scyl2_ST ●Srpk1_ST ●

Pim3_ST ●

Plk3_ST ●

Pim2_ST ●

Camk1d_ST ●

Snrk_ST ●

Chek2_ST ●

Vrk3_ST ●

Nek2_ST ●

Cdk18_ST ●

Limk2_ST ●

Csnk1d_ST ●

Vrk1_ST ●

Araf_ST ●

Clk3_ST●

Prkaa2_ST

●

Csnk1a1_ST

●

Cdk9_ST

●

Nek4_ST

●

Mapkapk3_ST

●

Nek8_ST

●

Fastk_ST

●

Mknk2_ST

●

Dyrk2_ST

●

Mapk6_ST

●

Cdk1_ST

Peak phases of TF accumulations vs. those of predicted motifs