MYC/MIZ1-dependent gene repression inversely coordinates the circadian clock with cell cycle and proliferation

Abstract

The circadian clock and the cell cycle are major cellular systems that organize global physiology in temporal fashion. It seems conceivable that the potentially conflicting programs are coordinated. We show here that overexpression of MYC in U2OS cells attenuates the clock and conversely promotes cell proliferation while downregulation of MYC strengthens the clock and reduces proliferation. Inhibition of the circadian clock is crucially dependent on the formation of repressive complexes of MYC with MIZ1 and subsequent downregulation of the core clock genes BMAL1 (ARNTL), CLOCK and NPAS2. We show furthermore that BMAL1 expression levels correlate inversely with MYC levels in 102 human lymphomas. Our data suggest that MYC acts as a master coordinator that inversely modulates the impact of cell cycle and circadian clock on gene expression.

ARTICLE

Received 3 Nov 2015 | Accepted 29 Apr 2016 | Published 24 Jun 2016

Anton Shostak1, Bianca Ruppert1, Nati Ha1, Philipp Bruns2, Umut H. Toprak2, ICGC MMML-Seq Projectw, Roland Eils2,3, Matthias Schlesner2, Axel Diernfellner1 & Michael Brunner1

The circadian clock and the cell cycle are major cellular systems that organize global physiology in temporal fashion. It seems conceivable that the potentially conicting programs are coordinated. We show here that overexpression of MYC in U2OS cells attenuates the clock and conversely promotes cell proliferation while downregulation of MYC strengthens the clock and reduces proliferation. Inhibition of the circadian clock is crucially dependent on the formation of repressive complexes of MYC with MIZ1 and subsequent downregulation of the core clock genes BMAL1 (ARNTL), CLOCK and NPAS2. We show furthermore that BMAL1 expression levels correlate inversely with MYC levels in 102 human lymphomas. Our data suggest that MYC acts as a master coordinator that inversely modulates the impact of cell cycle and circadian clock on gene expression.

DOI: 10.1038/ncomms11807 OPEN

MYC/MIZ1-dependent gene repression inversely coordinates the circadian clock with cell cycle and proliferation

1 Heidelberg University, Biochemistry Center, Im Neuenheimer Feld 328, D-69120 Heidelberg, Germany. 2 Division Theoretical Bioinformatics (B080), German Cancer Research Center (DKFZ), Im Neuenheimer Feld 280, D-69120 Heidelberg, Germany. 3 Department for Bioinformatics and Functional Genomics, Institute for Pharmacy and Molecular Biotechnology (IPMB) and BioQuant, Heidelberg University, Im Neuenheimer Feld 364, D-69120 Heidelberg, Germany. Correspondence and requests for materials should be addressed to M.B. (email mailto:[email protected]

Web End [email protected] ).

wA full list of consortia members appear at the end of the paper.

NATURE COMMUNICATIONS | 7:11807 | DOI: 10.1038/ncomms11807 | http://www.nature.com/naturecommunications

Web End =www.nature.com/naturecommunications 1

ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/ncomms11807

Many aspects of mammalian physiology and behaviour are rhythmically regulated by the circadian clock1. On a cellular level, the circadian clock is dependent on

interconnected transcriptional/translational feedback loops. In brief, the core transcription activator complex BMAL1/CLOCK (or its homologue BMAL1/NPAS2) rhythmically activates expression of clock genes including CRYs, PERs, REV-ERBs and RORs. CRYs and PERs are inhibitors of CLOCK/BMAL, whereas REV-ERBs are repressors that control in coordination with ROR activators expression of BMAL1, CLOCK and NPAS2. The D-box-specic transcription factors E4BP, DBP, TEF and HLF additionally contribute to the regulation of specic clock genes2.

Disruption or misalignment of circadian rhythms in humans has been associated with numerous pathological conditions including cancer3,4. Mice with chronic jet-lag or disrupted circadian physiology due to lesions of their suprachiasmatic nuclei exhibit accelerated growth of tumours5,6. Vice versa, cancer types correlating with impaired cell cycle and proliferation frequently exhibit aberrant expression of clock genes7,8. MYC is an oncogene, which is found to be deregulated in different cancers and, amplication of MYC often correlates with tumour aggression and poor prognosis9. MYC and its partner MAX are, like the circadian transcription factors BMAL1, CLOCK and, NPAS2, members of the bHLH transcription factors family, which form heterodimers that bind to so-called E-box motifs. MYC regulates transcription of up to 15% of the transcriptome including genes involved in apoptosis, cell growth and proliferation10,11. Recently, MYC has been suggested to attenuate the circadian clock by activating via circadian E-box sites transcription and expression of REV-ERBa/b, which would then repress transcription of BMAL1 (ref. 12). Since the DNA-binding specicity of MYC/MAX and CLOCK/BMAL1 complexes is highly similar, it seems conceivable that overexpressed MYC could constitutively activate and overexpress the E-box-dependent circadian repressor genes REV-ERBa/b, PER1/2 and CRY1/2.

Here we show, that overexpression of MYC attenuates the circadian clock of U2OS cells. Downregulation of the clock by overexpressed MYC is dependent on MYC/MIZ1 complexes, which are recruited to non-E-box sites in the promoters of BMAL1 and CLOCK. MYC/MIZ1 complexes stimulate proliferation of U2OS cells, suggesting that MYC inversely correlates the circadian clock and the cell cycle.

ResultsBMAL1 and MYC share common target genes. We therefore tested whether and how MYC/MAX and CLOCK/BMAL1 might regulate common circadian genes. A comparison of published chromatin immunoprecipitation sequencing (ChIP-seq) data sets of U2OS cells revealed a signicant overlap between both cistromes, with 28% (574/2048) of BMAL1 binding sites overlapping with sites also bound by MYC (refs 13,14; Fig. 1a). Such common binding sites include the core circadian clock genes REV-ERBa,

PER1/2 and DBP as well as clock-controlled genes such as SCN5A (Fig. 1b and Supplementary Fig. 1a). However, co-transfection of HEK293 cells with MYC/MAX expressing constructs did, in contrast to CLOCK/BMAL1, not strongly activate the circadian reporter genes REVERBa-luc, PER2-luc and SCN5A-luc (Fig. 1c).

To compare the activating potential of MYC/MAX and CLOCK/BMAL1 at E-boxes we assayed expression of a minimal promoter fused to 6 synthetic E-box elements (6xEbox-luc). Similar synthetic E-box reporters have been shown to be activated by MYC (ref. 15). Co-transfection of the 6xEbox-luc reporter with CLOCK and BMAL1 vectors resulted in notably higher luciferase activity than co-transfection with MYC and MAX vectors (14 fold

versus 34 fold; Fig. 1d). Interestingly, simultaneous expression of MYC/MAX together with CLOCK/BMAL1 hampered activation of the 6xEbox-luc reporter (Fig. 1d). Similarly, MYC/MAX interfered with stronger activation of REVERBa-luc and PER2-luc reporter genes by CLOCK/BMAL1 (Supplementary Fig. 1b). The data suggest that MYC/MAX has a weaker activation potential than CLOCK/BMAL1 at synthetic as well as endogenous circadian promoters. Yet, MYC/MAX is functionally dominant over CLOCK/BMAL1.

Overexpression of MYC disrupts the circadian clock. Next, we generated a U2OS cell line expressing a doxycycline-inducible V5-tagged MYC (U2OS t-rex tetO-MYC:V5). In presence of doxycycline total MYC (endogenous induced) was B9-fold

overexpressed (Fig. 4d and Supplementary Fig. 4c).

Twenty four and thirty six hours after induction, MYC:V5 was efciently recruited to circadian E-box sites in REV-ERBa and

PER2 (Fig. 1e). Rhythmic recruitment of BMAL1 to these loci was not compromised, yet BMAL1 occupancy was reduced 36 h after induction of MYC:V5 (Fig. 1f). The data suggest that at any given time the saturation level of the E-boxes with either transcription factor was rather low such that the transcription factors did not physically compete for common binding sites. The functional dominance of MYC/MAX could reect a MYC/MAX induced chromatin state that allows binding of CLOCK/BMAL1 but interferes with stronger activation of target genes.

We then asked whether overexpression of MYC affects expression levels and circadian rhythms of clock genes. Induction of transgenic MYC:V5 attenuated the circadian expression rhythms of PER2-luc and Bmal1-luc reporters in synchronized U2OS cells, while expression of green uorescent protein (control) had no effect (Fig. 1g and Supplementary Fig. 1c,d). Unexpectedly, however, the expression level and rhythm of the non-E-box-dependent Bmal1-luc reporter were strongly attenuated already shortly after induction of MYC:V5, whereas rhythmic expression of the E-box regulated PER2-luc reporter was affected with delayed kinetics (Fig. 1g). Surprisingly, expression levels of PER2-luc decreased in the presence of overexpressed MYC (Supplementary Fig. 1c) indicating that the MYC:V5 did not activate the E-box containing circadian promoter. Overexpression of MYC:V5 attenuated expression of endogenous BMAL1 and blunted its circadian prole about 1 day earlier and more strongly than the rhythm of the E-box containing PER2 gene (Fig. 1g,h). Furthermore, MYC:V5 expression caused downregulation of the non-E-box genes CLOCK and NPAS2 (Supplementary Fig. 1e).

It has been suggested that MYC activates REV-ERBa via E-boxes, which in turn would downregulate BMAL1 and thereby attenuate the circadian clock12. Indeed, overexpression of a doxycycline-inducible FLAG-tagged REV-ERBa repressed

Bmal1-luc levels and damped its rhythms (Supplementary Fig. 2a), indicating that REV-ERBa is a potent repressor of

BMAL1 in U2OS cells.

We therefore asked whether MYC activates REV-ERBa in U2OS cells. Circadian expression proles revealed that levels of REV-ERBa, as well as of REV-ERBb, were downregulated rather than activated by overexpressed MYC (Fig. 2a). The data are in agreement with the dominant negative effect of MYC/MAX over CLOCK/BMAL1 in transient expression of REVERBa-luc (Supplementary Fig. 1b).

To analyse the impact of MYC on REV-ERBa in more detail we constructed a MYC-overexpressing cell line (U2OS t-rex tetO-MYC:V5) that expresses under control of the Rev-erba promoter a destabilized nuclear venus reporter (Rev-VNP)16.

Rev-VNP was rhythmically expressed as assessed by quantifying

2 NATURE COMMUNICATIONS | 7:11807 | DOI: 10.1038/ncomms11807 | http://www.nature.com/naturecommunications

Web End =www.nature.com/naturecommunications

NATURE COMMUNICATIONS | DOI: 10.1038/ncomms11807 ARTICLE

the total uorescence and the number of uorescent objects (cells) per well (Fig. 2b, black curves). Overexpression of MYC triggered a transient activation of Rev-VNP with a peak B7 h after induction, which was followed by downregulation and blunting of the circadian Rev-VNP rhythm (Fig. 2b, red curves).

We then integrated over a 3-day time course (2496 h) the total number of uorescent objects (cells expressing Rev-VNP above threshold) to assess whether the MYC-dependent loss of rhythmicity in the population of cells was due to desynchronization (same number of uorescent objects in presence and absence of MYC) or due to an attenuated rhythm, that is, a reduction of the number of highly rhythmic cells (less uorescent objects above threshold). One day after induction of MYC:V5, the average number of uorescent objects was B2-fold reduced, indicating a decrease of the number of highly rhythmic cells rather than desynchronization of the population of cells (Fig. 2b, inset). MYC-dependent attenuation of the circadian Rev-VNP rhythm rather than MYC-induced desynchronization was also obvious on the level of single cells (Supplementary Movie 1). The data suggest that induction of MYC:V5 does not trigger a sustained overexpression of REV-ERBa but rather reduces its expression level and rhythm. Corresponding results were obtained with a REVERBa-luc reporter (Supplementary

Fig. 2b).

Altman et al.12 reported induction of REV-ERBa in response to activation of MYC. However, they quantied REV-ERBa at a

single time point early after induction of MYC (24 h). At this time REV-ERBa levels may still be slightly elevated due to its transient induction peaking after 7 h (Fig. 2b).

We thus knocked down REV-ERBa by short interfering RNA (siRNA) to assess whether the MYC-induced downregulation of BMAL1 was dependent on REV-ERBa. REV-ERBa was B3-fold downregulated by specic versus control siRNA (Fig. 2c). When MYC was overexpressed, REV-ERBa levels were additionally reduced by 40% in presence of REV-ERBa specic siRNA and by 25% in presence of control siRNA (Fig. 2c). Downregulation of REV-ERBa by siRNA resulted in elevated expression of BMAL1 (Fig. 2c), indicating that REV-ERBa is a potent repressor of BMAL1 in U2OS cells. However, overexpressed MYC repressed BMAL1 in both, REV-ERBa depleted and control cells (62 and 69%, respectively), indicating that MYC acted independent of REV-ERBa. Corresponding results were obtained with a

Bmal1-luc reporter (Fig. 2d).

Downregulation of the positive limb of the circadian feedback loop has been shown to be sufcient to disrupt the molecular clock in U2OS cells17. Accordingly, the MYC-dependent attenuation of expression level and circadian rhythm of Bmal1-luc was partially rescued by constitutive overexpression of BMAL1 together with CLOCK (Supplementary Fig. 2c,d). Together, these data suggest that overexpression of MYC did not enhance expression of REV-ERBa in a sustained manner. Rather, overexpressed MYC compromised the circadian system primarily

a

b

1 kb

1 kb

MYC binding sites

BMAL1 binding sites

BMAL1 BS CLOCK BS MYC BS Induced MYC BS

BMAL1 BS

CLOCK BS

MYC BS Induced MYC BS

19,440 574 1,474

hPER2

hREV-ERB

c d

g

PER2-luc

Bmal1-luc

8 REVERB-luc

PER2-luc SCN5-luc GAPDH-luc

20 6xEbox-luc

+

+ +

+ +

10,000

Relative expression (a.u.)

Relative expression (a.u.)

*

10,000

*

Bioluminescence (cps)

6

15

5,000

*

5,000

4

*

10

0

0

2

*

5

5,000

5,000

0

0

CLOCK+BMAL1 +

+

Ctrl MYC ox

pcDNA

+

+

+

+

pcDNA

+

10,000

10,000

+

+

+

MYC + MAX

+

+

+

CLOCK+BMAL1 MYC

MAX

+

0 24 48 72 96

0 24 48 72 96

30 30

5 5

30 30

30

30

Time after synchronization (h)

e f

h

PER2

BMAL1

PER2 REV-ERB Neg region

PER2 REV-ERB Neg region

1.5

Relative expression (a.u.)

1.0

30

2.0 *

MYC:V5 ChIP (% of input)

BMAL1 ChIP (% of input)

* *

0.8

*

1.5

20

1.0

*

0.6

1.0

0.4

10

0.5

0.5

0.2

Ctrl

MYC ox

0.0

0.0

0.0

24 36 24 36 24 36

0 24 36 24 36 24 36

24 30 36 42 48 54 60 66

24 30 36 42 48 54 60 66

Time after synchronization (h)

Time after synchronization (h)

Time after synchronization (h)

Figure 1 | Overexpression of MYC attenuates the circadian clock. (a) Overlap between native MYC (ref. 14) and BMAL1 (ref. 13) binding sites in U2OS cells. (b) PER2 and REV-ERBa loci with binding sites (BS) of BMAL1, CLOCK, native MYC and overexpressed MYC in U2OS cells (based on the data from refs 13,14). (c) MYC and MAX do not substantially induce the BMAL1/CLOCK target genes REV-ERBa, PER2 and SCN5a. HEK293 cells were transfected with

MYC, MAX, BMAL1 and CLOCK encoding plasmids (30 ng) together with the indicated circadian promoter-luc reporter plasmids. GAPDH-luc was transfected as a negative control (n 3). (d) MYC/MAX restricts stronger induction of 6xEbox-luc by CLOCK/BMAL1. HEK293 cells were transfected with 30 ng of

each BMAL1 and CLOCK plasmids, and with the indicated amounts (in ng) of MYC and MAX vectors (n 3). ChIP-PCR analysis of (e) MYC:V5 and

(f) BMAL1 binding to circadian E-boxes in PER2 and REV-ERBa promoters in synchronized and doxycycline-induced U2OS t-rex tetO-MYC:V5 cells (n 3).

(g) Bioluminescence recorded from synchronized Bmal1-luc and PER2-luc U2OS t-rex tetO-MYC:V5 cells (n 3). (h) Quantitative PCR (qPCR) analysis of

circadian expression proles of PER2 and BMAL1 transcripts in synchronized U2OS t-rex tetO-MYC:V5 cells (n 3). Data are presented as means.e.m.

*Po0.05; one-way (c,d) and two-way (e,h) analysis of variance (ANOVA) with Bonferroni post-test.

NATURE COMMUNICATIONS | 7:11807 | DOI: 10.1038/ncomms11807 | http://www.nature.com/naturecommunications

Web End =www.nature.com/naturecommunications 3

ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/ncomms11807

tetO-MYC:V5

a

1.5

REV-ERB[afii9825] REV-ERB[afii9826]

1.5

Relative expression (a.u.)

*

1.0

*

1.0

*

*

*

0.5

0.5

0.0

0.0

24 30 36 42 48 54 60 66

24 30 36 42 48 54 60 66

Time after synchronization (h)

Ctrl MYC ox

b

Rev-VNP tetO-MYC:V5

Ctrl MYC ox

Rel. fluor. objects

500,000

2,000

1.0

Rev-VNP fluorescence (a.u.)

Fluorescent objects per well

0.5

400,000

1,500

0.0

300,000

Ctrl

MYC ox

1,000

200,000

500

100,000

0 0 24 48 72 96

0

0 24 48 72 96

Time after synchronization (h)

c

tetO-MYC:V5 d

tetO-MYC:V5 Bmal1-luc

REV-ERB[afii9825] BMAL1

1.5

2.5

800,000

*

*

*

*

*

*

2.0

Relative expression (a.u.)

Integrated bioluminescence

600,000

1.0

1.5

62%

400,000

1.0

*

69%

*

0.5

*

200,000

0.5

0.0

0.0

0

REV-ERB[afii9825]

Neg siRNA siRNA

REV-ERB[afii9825]

Neg siRNA siRNA

REV-ERB[afii9825]

Neg siRNA siRNA

Ctrl

MYC ox

Figure 2 | MYC represses BMAL1 independent of REV-ERBa. (a) Quantitative PCR (qPCR) analysis of circadian expression proles of REV-ERBa and REV-ERBb transcripts in synchronized U2OS t-rex tetO-MYC:V5 cells (n 3). (b) Total uorescence and uorescent objects quantied from synchronized

U2OS t-rex Rev-VNP tetO-MYC:V5 cells (n 1). Inset: normalized number of uorescent objects (2496 h integration; n 2). (c) qPCR analysis of REV-ERBa

and BMAL1 transcripts in siRNA-transfected U2OS t-rex tetO-MYC:V5 cells (n 3). Unsynchronized cells were collected 24 h after doxycycline induction.

(d) Integrated bioluminescence of Bmal1-luc transfected U2OS t-rex tetO-MYC:V5 cells. Cells were transfected with indicated siRNAs 2 days before synchronization. Average expression levels (area under the curve) were determined over 72 h (n 3). Data are presented as means.e.m. *Po0.05;

two-way analysis of variance (ANOVA) with Bonferroni post-test.

via downregulation of BMAL1, CLOCK and its homologue NPAS2 by a pathway independent of REV-ERBa and REV-ERBb.

MYC represses the circadian clock via interaction with MIZ1. MYC is not only a transcription activator. When expressed at high level, MYC has the potential to form a repressive complex with MIZ1 (ZBTB17) and downregulate in E-box-independent fashion expression of MIZ1 target genes such as the cyclin-dependent kinase inhibitor genes p15 and p21 (refs 14,18,19).

We followed the MYC-dependent temporal expression proles of luciferase reporter constructs of established and putative MIZ1 target genes. After induction of MYC:V5 with doxycycline, Bmal1-luc was repressed with similar kinetics as p15-luc and p21-luc, while 6xEbox-luc was, as expected, induced when MYC was overexpressed (Fig. 3a). Examining a published ChIP-seq analysis of MIZ1 in U2OS cells14, we realized that MIZ1 binds to the promoters of BMAL1, CLOCK and NPAS2 (Fig. 3b, left panel). We therefore performed a ChIP-PCR analysis and found that

4 NATURE COMMUNICATIONS | 7:11807 | DOI: 10.1038/ncomms11807 | http://www.nature.com/naturecommunications

Web End =www.nature.com/naturecommunications

NATURE COMMUNICATIONS | DOI: 10.1038/ncomms11807 ARTICLE

a

tetO-MYC:V5 tetO-MYC:V5

b

MYC:V5 ChIP (% of input)

p21-luc

3,000

Bioluminescence (cps)

4,000

20,000

10 kb

400,000

6xEbox-luc

MIZ1 BS

0.6

24 h

3,000

15,000

36 h

2,000

10 kb

NPAS2

0.4

2,000

10,000

200,000

MIZ1 BS

1,000

1,000

5,000

0.2

Ctrl MYC ox

10 kb CLOCK

MIZ1 BS

0.0

0 0 5 10 15 20

p15-luc

0 0 5 10 15 20

Bmal1-luc

0 0 5 10 15 20

0

0 5 10 15 20

BMAL1

TSS NPAS2

Enhancer

TSSCLOCK Neg region

TSS

BMAL1

Time after Dox addition (h)

Neg siRNA MIZ1 siRNA#1

c d

tetO-MYC:V5 Bmal1-lucNeg siRNA MIZ1 siRNA#1 MIZ1 siRNA#2

1.5

4,000

4,000

4,000

Relative expression (a.u.)

Bioluminescence (cps)

3,000

3,000

3,000

1.0

*

*

*

2,000

2,000

2,000

0.5

1,000

1,000

1,000

0.0

0

0

0

MIZ1

NPAS2 CLOCK BMAL1

0 24 48 72

0 24 48 72

0 24 48 72

Time after synchronization (h)

Figure 3 | MYC directly represses clock genes via MIZ1. (a) Kinetics of MYC-dependent repression of p21-luc, p15-luc and Bmal1-luc and activation of 6xEbox-luc (n 3). Unsynchronized U2OS t-rex tetO-MYC:V5 cells were transiently transfected with the indicated reporter plasmids. At time point 0

MYC:V5 expression was either induced with doxycycline (MYC ox) or not induced with PBS (Ctrl). (b) Left panel: schematic of MIZ1 binding sites in NPAS2, CLOCK and BMAL1 based on data from Walz et al.14. Black triangles indicate regions amplied in ChIP-PCR analysis. Right panel: ChIP-PCR analysis showing recruitment of induced MYC:V5 to MIZ1 binding sites in NPAS2, CLOCK and BMAL1 at 24 and 36 h after synchronization of U2OS t-rex tetO-MYC:V5 cells (n 3). (c) Downregulation of MIZ1 with siRNA reduces expression of NPAS2, CLOCK and BMAL1 in U2OS cells (n 3). Transcript levels of the indicated

genes were determined by quantitative PCR (qPCR). (d) MYC-induced attenuation of the circadian clock is rescued by downregulation of MIZ1 (n 3).

Synchronized U2OS t-rex tetO-MYC:V5 Bmal1-luc cells were transfected with MIZ1 or control siRNAs. Note that downregulation of MIZ1 attenuates the circadian rhythm of Bmal1-luc and lengthens the period (Supplementary Fig. 3a,b). Data are presented as means.e.m. *Po0.05; Students t-test.

overexpressed MYC:V5 was recruited to the MIZ1 sites in BMAL1, CLOCK and NPAS2 (Fig. 3b, right panel), suggesting that MYC binds to and represses these genes via MIZ1.

When MYC is not overexpressed MIZ1 acts as a transcription activator20. To assess whether MIZ1 supports expression of BMAL1, CLOCK and NPAS2 in U2OS cells, we depleted MIZ1 with siRNA (Fig. 3c). Expression of BMAL1, CLOCK and NPAS2 was reduced indicating that MIZ1 contributes to the transcriptional activation of these clock genes. When MIZ1 was depleted by siRNAs, the circadian expression rhythm of Bmal1-luc was weakened in comparison to a treatment with control siRNA (Fig. 3d, black curves) and the circadian period was lengthened (Supplementary Fig. 3a). To assess whether the MYC-dependent disruption of the clock requires MIZ1, we induced MYC:V5 in MIZ1-depleted and control-treated cells (Fig. 3d, red curves). As expected, the circadian expression rhythm of Bmal1-luc was abolished when MYC:V5 was overexpressed in control-treated cells (Fig. 3d, red curve left panel). Concomitant downregulation of MIZ1, however, partially restored the expression level and circadian rhythms of Bmal1-luc despite overexpression of MYC:V5 (Fig. 3d, red curves middle and right panels). Correspondingly, the rhythm of 6xEbox-luc was severely attenuated by induced MYC:V5 and fully rescued by siRNA depletion of MIZ1 (Supplementary Fig 3b). Together the data indicate that overexpressed MYC represses BMAL1 directly via MIZ1-dependent recruitment of MYC to MIZ1 binding sites.

MYC-mediated attenuation of the clock requires MIZ1. To challenge this hypothesis, we generated the V5-tagged MYC V394D mutant, which is impaired in its interaction with MIZ1 (ref. 21) and in addition the variant V393D. ChIP-PCR analysis demonstrated that, as was shown for V394D21, the newly generated V393D variant binds with similar afnity to E-boxes as WT MYC (Supplementary Fig. 4a). Both mutant versions of MYC:V5 induced expression of 6xEbox-luc in HEK293 cells and interfered in dominant negative fashion with

CLOCK/BMAL1-dependent activation of 6xEbox-luc to the same extent as wild-type MYC:V5 (Fig. 4a). Theses data suggest that the E-box-dependent functionality of MYC:V5 V394D and MYC:V5 V393D was not affected. Both mutant versions interacted with MAX with similar efciency as MYC:V5 (Fig. 4b, upper panel; and Supplementary Fig. 4b). In contrast, the V394D and V393D variants showed markedly reduced association with MIZ1 (Fig. 4b, lower panel; and Supplementary Fig. 4b). Next, we produced stable U2OS cell lines harbouring doxycycline-inducible MYC:V5 V394D and MYC:V5 V393D genes. Both inducible transcripts accumulated to similar levels as MYC:V5 (Supplementary Fig. 4c). However, repression of p15-luc, p21-luc as well as Bmal1-luc reporters was alleviated in the cell lines expressing the V393D and V394D versions of MYC:V5 (Fig. 4c). The levels of CLOCK and BMAL1 proteins were strongly reduced by overexpression of MYC:V5, but remained essentially unaffected by induction of MYC:V5 V393D (Fig. 4d). Circadian expression rhythms of Bmal1-luc and 6xEbox-luc reporters were strongly attenuated by overexpression of wild-type MYC:V5, but only mildly affected by overexpression of V394D and essentially unaffected by the V393D version (Fig. 4e and Supplementary Fig. 4d). Similarly, Rev-VNP (ref. 16) expression levels and rhythms were strongly attenuated by overexpressed MYC:V5 but unaffected by induction of MYC:V5 V393D (Fig. 4f and Supplementary Movies 1 and 2). The data demonstrate that overexpressed MYC attenuates the circadian clock via MIZ1-mediated E-box-independent repression.

MYC-dependent cell cycle stimulation requires MIZ1. Elevated MYC expression is considered to drive cell growth and proliferation through genome-wide interference with physiological regulation of E-box-dependent transcription22. Since the E-box-dependent activation properties are preserved in the V393D version of MYC:V5 (Fig. 4a), we asked how it affects cell growth and proliferation. Consistent with previous studies21,

NATURE COMMUNICATIONS | 7:11807 | DOI: 10.1038/ncomms11807 | http://www.nature.com/naturecommunications

Web End =www.nature.com/naturecommunications 5

ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/ncomms11807

a b c

6xEbox-luc

10

4 *

FLAG-MAX MYC:V5

*

*

kDa 72

WT V394D V393D

1.0

* *

8

Relative expression (a.u.)

Relative bioluminescence (a.u.)

3

WB: V5

*

IP: V5

MYC:V5 WT

MYC:V5 V394D

MYC:V5 V393D

0.8

*

2

Relative expression (a.u.)

6

26

*

*

WB: FLAG

0.6

*

4

0.4

1

FLAG-MIZ1 MYC:V5

2

kDa 72

WT V394D V393D

0.2

0

0

WB: V5

IP: V5

0.0

p15-luc

p21-luc

Bmal1-luc

WB: FLAG

95

Bmal1-luc

d

e

1.5

Relative protein abundance (a.u.)

BMAL1

*

tetO-MYC:V5 WT

tetO-MYC:V5 V394D

tetO-MYC:V5 V393D

MYC:V5 V393D

MYC:V5 WT

Ctrl MYC ox Ctrl MYC ox

6,000

1,500

6000

1.0

Ctrl MYC ox

kDa

72

72

55

Bioluminescence (cps)

4,000

1,000

4000

CLOCK

TUB

0.5

95

2,000

500

2000

BMAL1

0.0

1.5 CLOCK

*

0

0

0

MYC

1.0

2,000

500

2000

0.5

4,000

1,000

4000

0.0

0 24 48 72 96 120

0 24 48 72 96 120

0 24 48 72 96 120

Time after synchronization (h)

Ctrl MYC ox

tetO-MYC:V5 WT tetO-MYC:V5 V393D

f

Ctrl MYC ox

800,000

4,000

10,000

Rev-VNP fluorescence (a.u.)

Fluorescent objects per well

8,000

400,000

2,000

Rev-VNP fluorescence (a.u.)

Fluorescent objects per well

600,000

3,000

1,000,000

6,000

500,000

4,000

200,000

1,000

2,000

0

0

0

0 24 48 72 96

0 24 48 72 96

0 24 48 72 96

0

0 24 48 72 96

Time after synchronization (h)

Figure 4 | MYC mutants compromised in MIZ1 interaction do not disrupt the circadian clock. (a) Transactivation assay in HEK293 cells showing that MYC:V5 WT, V394D and V393D equally induce 6xEbox-luc expression and compete with CLOCK/BMAL1 in a dominant negative manner (n 3).

(b) Anti-V5 immunoprecipitation of MYC:V5 versions in HEK293 lysates showing that MYC:V5 V394D and V393D interact with MAX (upper panel) but not with MIZ1 (lower panel). Co-immunoprecipitation (Co-IP) of FLAG-tagged MAX and MIZ1 was detected with anti-FLAG antibodies. Refer to Supplementary Fig. 4a for inputs and reciprocal anti-FLAG co-IPs. (c) Overexpressed MYC:V5 V394D and V393D are inefcient in repression of MIZ1 target genes. Expression of transiently transfected p15-luc, p21-luc and Bmal1-luc reporters in U2OS t-rex tetO-MYC:V5 cells expressing the indicated versions of MYC. Bioluminescence was quantied 18 h after MYC induction with doxycycline and normalized to PBS-treated samples (n 3). (d) Western blot

analysis (left) and densitometric protein quantication (right) of CLOCK and BMAL1 in U2OS cells overexpressing MYC and MYC V393D 24 h after doxycycline induction (n 3). (e) Baseline-subtracted bioluminescent traces from U2OS t-rex tetO-MYC:V5 WT, V394D and V393D cells transiently

transfected with Bmal1-luc (n 3). For the raw data refer to Supplementary Fig. 4d. (f) Total uorescence and uorescent objects quantied from

synchronized U2OS t-rex Rev-VNP cells stably transfected with inducible MYC:V5 WT and V393D (n 1). Data are presented as means.e.m. *Po0.05;

one-way analysis of variance (ANOVA) with Bonferroni post-test.

induction of MYC:V5 reduced the fraction of cells in G1 phase and increased the proportion of cells in S, G2 and M phases (Fig. 5a,b). In contrast, overexpression of MYC:V5 V393D did not support proliferation to a substantial extent. These data suggest that stimulation of proliferation of U2OS cells by MYC is crucially dependent on gene repression via MIZ1. Together the data suggest that overexpression of MYC supports proliferation and attenuates the circadian clock predominantly via MIZ1.

Knockdown of MYC induces circadian amplitude. Next, we analysed effects of MYC depletion. Knockdown of MYC by siRNA resulted in decreased proliferation of U2OS cells (Fig. 5c,d). Concomitantly, expression levels of BMAL1 and CLOCK increased (Fig. 5e). Furthermore, the relative amplitudes (the oscillation amplitude divided by the mean expression) of the circadian expression rhythms of Bmal1-luc and other circadian reporters signicantly improved in cells transfected with MYC siRNA (Fig. 5f and Supplementary Fig. 5a). Corresponding effects

were observed in U2OS cells expressing Rev-VNP (Supplementary Fig. 5b, Supplementary Movie 3). Thus, downregulation of MYC with siRNA increased the relative amplitude of the Rev-VNP rhythm. In summary, the data indicate that downregulation of MYC attenuates proliferation and strengthens the circadian clock.

Inverse correlation of MYC and clock genes in lymphoma. MYC is overexpressed in many tumours and depletion of MYC is known to detain tumour growth22. Encouraged by the inverse correlation of MYC with BMAL1 and CLOCK expression in U2OS cells, we assessed expression of these genes in human lymphoma by analysing RNA-seq data of 102 patient samples. Lymphoma was chosen since this type of malignancy requires interaction of MYC and MIZ1 (ref. 23). Consistent with our observation in U2OS cells, expression of both clock genes correlated inversely with MYC levels in human lymphoma (Pearson coefcient 0.61 and 0.37 for BMAL1 and

CLOCK, respectively; n 102), whereas MYC-activated genes

6 NATURE COMMUNICATIONS | 7:11807 | DOI: 10.1038/ncomms11807 | http://www.nature.com/naturecommunications

Web End =www.nature.com/naturecommunications

NATURE COMMUNICATIONS | DOI: 10.1038/ncomms11807 ARTICLE

a b

Cells in G0-G1 phase (FUCCI-Red expressing cells)

G0-G1 S G2-M

Ctrl

*

MYC ox

20

10

20

5,000

*

MYC-induced change

in cell cycle stage (%)

MYC:V5 V393DMYC:V5 WT

G0-G1 cells per well

4,000

3,000

0

2,000

10

1,000

MYC:V5 WT

MYC:V5 V393D

MYC:V5 WT MYC:V5 V393D

*

Ctrl MYC ox

0

c e

d

BMAL1

CLOCK

U2OS

U2OS

*

*

30

Relative protein abundance (a.u.)

MYC siRNA

2

1,000

MYC siRNA-induced change

in cell cycle stage (%)

Relative cell number (%)

20

*

WB:

1

800

CLOCK

TUB

Neg siRNA

kDa

72 72

55

10

95

600

0

BMAL1

0

400

MYC

6 *

10

*

200

Neg siRNA MYC siRNA

4

*

Neg siRNA MYC siRNA

2

20 G0-G1 S G2-M

0 0 1 2 3 4 5Time (days)

0

f

Bmal1-luc

PER2-luc ATG3-luc EIF5A2-luc PLD6-luc

SCN5[afii9825]-luc

1.5

Relative amplitude (a.u.)

1.5 *

0.25

0.20 *

0.25

0.4

*

*

*

*

0.20

0.20

0.15

0.3

1.0

1.0

0.15

0.15

0.10

0.2

0.10

0.10

0.5

0.5

0.05

0.1

0.05

0.05

0.0

0.0

0.00

0.00

0.00

0.0

Neg siRNA MYC siRNA

Figure 5 | MYC inversely regulates the circadian clock and proliferation. (a) MYC:V5 WTand V393D U2OS cells were stained with propidium iodide 48 h after induction and DNA content was quantied by FACS (n 3). Values indicate difference (in %) to PBS-treated cells in the respective cell cycle phase.

(b) Left panel: uorescence microscopy of U2OS t-rex tetO-MYC:V5 cells stably expressing mCherry-Cdt1 (FUCCI-Red G1 marker) 48 h after treatment with doxycycline to induce MYC:V5 (MYC ox) or PBS (Ctrl). Scale bar, 300 mm. Right panel: quantication of cells in G0 or G1 phase by mCherry-Cdt1 expression (n 3) (c) FACS analysis of U2OS cells stained with propidium iodide 48 h after transfection with MYC siRNA (n 3). Values indicate

difference to cells transfected with negative siRNA. (d) Growth curve of U2OS cells transfected with MYC siRNA and negative siRNA (n 3). (e) Western

blot analysis (left) and densitometric protein quantication (right) showing that MYC depletion by siRNA supports increased BMAL1 and CLOCK expression in U2OS cells (n 3). (f) Relative amplitudes (ChronoStar software) of circadian luciferase rhythms of indicated U2OS reporter cell lines

(n 3). The cells were transfected with MYC siRNA and negative siRNA as indicated. Data are presented as means.e.m. *Po0.05; Students t-test (e,f),

one-way (b,c) and two-way (a,d) analysis of variance (ANOVA) with Bonferroni post-test.

such as NCL and NCLN showed a positive correlation (Fig. 6a,b and Supplementary Fig. 6). MYC levels did not correlate positively with expression of circadian genes regulated via E-boxes. Particularly, REV-ERBa and b did not signicantly correlate with MYC.

The circadian clock and cell cycle/proliferation are major programs controlling expression of specic, potentially overlapping sets of genes2427. Since the simultaneous regulatory activity of both programs in the same cell may create conicting signals, it seems conceivable that their activity is coordinated under physiological conditions. It has been shown that in regenerating mouse liver, the circadian clock gates the cell

cycle28 and that in NIH3T3 cells the cell cycle resets the circadian clock29,30. The circadian clock in the liver of a living animal oscillates with high amplitude, while the circadian clock in isolated cells appears to be substantially weaker13,31. Hence, the dominance of one cycling programme over the other may depend on their relative strength or rhythmic momentum. Our observations suggest that MYC is a master regulator coordinating both programs. Expression levels of MYC determine the relative strength of the circadian clock versus cell cycle/proliferation and hence, their impact on gene expression and cellular physiology. MYC attenuates the circadian system and promotes proliferation. Both functions are critically dependent on complexes of MYC

NATURE COMMUNICATIONS | 7:11807 | DOI: 10.1038/ncomms11807 | http://www.nature.com/naturecommunications

Web End =www.nature.com/naturecommunications 7

ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/ncomms11807

a

BMAL1

BMAL1

CLOCK

CRY1

CRY2

PER2

REV-ERB[afii9825]

REV-ERB[afii9826]

NCL

Gene expression (RPKM)

Gene expression (RPKM)

8

0

4

6

15

10

2

15

20

400

15

300

4

2

8

4

6

4

2

3

1

10

10

5

6

200

5

10

5

2

100

10 100 1,000

0 10 100 1,000

0 10 100 1,000

0 10 100 1,000

0 10 100 1,000

0 10 100 1,000

0 10 100 1,000

0 10 100 1,000

MYC expression (RPKM)

b

CLOCK CRY1

CRY2 NCL

050 50100 >100

PER2 REV-ERB[afii9825]

REV-ERB[afii9826]

8

0

n = 72 n = 15 n = 15 n = 72 n = 15 n = 15 n = 72 n = 15 n = 15 n = 72 n = 15 n = 15 n = 72 n = 15 n = 15 n = 72 n = 15 n = 15 n = 72 n = 15 n = 15 n = 72 n = 15 n = 15

*

*#

7

6

5

1

10

*

3.5

3.0

2.5

2.0

1.5

1.0

10

*

300

50

* #

6

5

1

250

3

6

5

4

3

2

4

8

2

*

8

6

*

*

4

6

4

200

* *

*#

2

3

4

150

2

4

*

2

2

100

050 50100 >100

050 50100 >100

050 50100 >100

050 50100 >100

050 50100 >100

050 50100 >100

050 50100 >100

c

High MYC

M S

G1

G2

Low MYC

129 3 6

Figure 6 | MYC inversely correlates with BMAL1 expression in human lymphomas. (a) Scatter plots of expression levels of MYC versus the indicated clock genes in 102 human lymphoma samples of the ICGC MMML-Seq project37 (RPKM: log2 sequence reads per kilobase transcript per million reads38). Expression of MYC versus its established target NUCLEOLIN (NCL) is shown as a positive control. (b) Data were binned according to the indicated MYC expression levels. Clock gene expression levels are shown by Box-plots. Signicant differences (Po0.05; Students t-test) of expression levels relative to box 1 (050) and box 2 (50100) are indicated by asterisks (*) and number signs (#), respectively. (c) Model of the coordinating function of MYC. Left panel: high levels of MYC suppress the circadian clock by MIZ1-dependent downregulation of BMAL1/CLOCK (see text), which results in low amplitude expression rhythms of clock-controlled genes. On the other hand, high MYC levels support cell growth and proliferation (for example, by inhibition of p15 and p21, see text). Right panel: at low levels of MYC, the circadian clock is not inhibited and supports high amplitude expression rhythms of clock-controlled genes. Low levels of MYC do not support cell growth and proliferation.

with MIZ1. Such repressive complexes form preferentially at high MYC levels32. The inverse correlation of BMAL1 versus MYC expression levels in human lymphomas and the absence of a positive correlation of MYC with circadian E-box genes support this hypothesis. The correlation suggests that circadian physiology might be compromised in tumours with amplied MYC. Strategies targeting the interaction of MYC with MIZ1 could help to recover circadian control over cellular physiology in malignant cells and potentially inhibit uncontrolled growth without major effects on differentiated post-mitotic cells with low MYC expression.

It seems conceivable that rhythmic downregulation of metabolic and biosynthetic functions by the circadian clock may be in conict with rapid cell growth and proliferation. Hence, aside from malignant conditions, MYC may also coordinate the relative impact on gene expression of the circadian clock versus cell cycle/proliferation under physiological conditions such as in the developing and differentiating embryo.

Methods

Cell culture and transfections. U2OS t-rex (T-Rex, Life Technologies) and HEK293T (ATCC) cells were maintained in DMEM supplemented with 10% fetal

bovine serum (FBS) and 1xPenStrep. Cell culture reagents were obtained from Life Technologies unless indicated differently. U2OStx were transfected with AhdI-linearized pcDNA4/TO vectors containing different alleles of human MYC using Xfect (Clontech) and stable transfectants were selected using growth medium supplemented with 50 mg ml 1 hygromycin and 100 mg ml 1 zeocin (InvivoGen).

Positive clones were subsequently, transiently or stably transfected with circadian reporter constructs using Xfect reagent and, if applicable, selected with 1 mg ml 1 puromycin (InvivoGen). U2OS stable cell lines expressing the promoter-luciferase constructs were described previously13. For RNAi experiments, U2OS cells were transfected with the indicated siRNA (sequences are given in Supplementary Table 1) using lipofectamine RNAiMAX reagent according to manufacturers protocols. For luciferase reporter assays, HEK293T cells were transfected with the indicated constructs using Lipofectamine2000 and 24 h later luciferase expression was assayed using Dual-luciferase Reporter Assay (Promega) and an EnSpire Reader (Perkin Elmer).

Plasmid constructs. Bmal1-luc and 6xEbox-luc vectors were kindly provided by Dr Steven A Brown and Dr Achim Kramer33,34. p15-luc, p21-luc and MIZ1 open reading frame (ORF) vectors were kindly provided by Dr Elmar Wolf. Rev-VNP uorescent reporter containing 13 kb of murine Rev-erba promoter coupled to

Venus-NLS-PEST was kindly provided by Dr Ueli Schibler. Promoter sequences (1 kb) of PER2, REV-ERBa, SCN5a and GAPDH genes cloned in pGL4.20puro vector were described previously13. BMAL1, CLOCK and MAX ORFs were amplied from U2OS complementary DNA (cDNA) and cloned in pcDNA4/TO vector. The MYC and FLAG:REV-ERBa ORFs were obtained from Addgene and subcloned in pcDNA4/TO. MYC V394D and V393D mutations were performed

8 NATURE COMMUNICATIONS | 7:11807 | DOI: 10.1038/ncomms11807 | http://www.nature.com/naturecommunications

Web End =www.nature.com/naturecommunications

NATURE COMMUNICATIONS | DOI: 10.1038/ncomms11807 ARTICLE

using DF-Pfu polymerase (Bioron). Cloning and mutagenesis primers are available on request.

Real-time bioluminescence monitoring. For RNAi or transfection experiments, cells were seeded into a 96-well plate (20,00030,000 cells per well) and next day transfected with the indicated siRNAs. 24 h later the transfection medium was removed and cells were synchronized with 1 mM dexamethasone for 20 min and washed with PBS. After addition of warm luminescene medium (DMEM w/o Phenolred (PAA) supplemented with 10% FBS, 25 mM Hepes, 1xPenStrep, and0.125 mM luciferin (BioSynth), 10ng ml 1 doxycycline) the plate was sealed and bioluminescence was recorded for 30 min intervals at 37 C with an EnSpire Reader (Perkin Elmer). Circadian period and amplitude analysis was performed using ChronoStar software35.

Gene expression analysis. Transfected or synchronized cells were lysed with TriFaster (GeneON) and total RNA was extracted according to manufacturers protocol. cDNA was synthesized with Maxima First Strand cDNA Synthesis Kit (Thermo Scientic). Quantitative PCR was performed using GoTaq Master Mix (Promega) and LightCycler 480 (Roche) and relative gene expression was quantied using a DDCt method with GAPDH as a reference gene. Primer sequences are listed in Supplementary Table 1.

Chromatin immunoprecipitation (ChIP). U2OS t-rex cells were collected 24 and 36 h after synchronization (10 ng ml 1 doxycycline added at time point 0) and immediately cross-linked in 1% formaldehyde for 10 min. ChIP was performed as described previously with minor modications36. In brief, pelleted nuclei were suspended in 300 ml IP buffer (150 mM NaCl, 5 mM EDTA, NP-40 (0.5%), Triton

X100 (1.0%), 50 mM Tris-HCl, pH 7.5) supplemented with 0.1% SDS and sonicated (30 s on/30 s off cycles for 20 min) using a Bioruptor (Diagenode Inc.). Sheared chromatin (equivalent of 106 cells) was incubated overnight at 4 C with 3 ml of anti-BMAL1 (ref. 13; 0.1 mg ml 1) and 0.5 ml anti-V5 antibodies(1 mg ml 1, 46-0705, Life Technologies), and obtained immune complexes were precipitated with salmon sperm DNA blocked protein A-agarose beads (Millipore)

equilibrated ve times with IP buffer. Precipitated DNA was recovered by boiling

for 10 min in 10% Chelex slurry (Bio-Rad) followed by Proteinase K (150 mg ml 1,

New England Biolabs) treatment at 55 C for 30 min. Proteinase K was subsequently inhibited by boiling at 95 C for 10 min and beads were removed by centrifugation at 12,000g at 4 C. DNA-containing supernatants were analysed by quantitative PCR, and values were normalized to percentage of input. Primer sequences are listed in Supplementary Table 1.

Cell proliferation assays. Prior to siRNA transfection, U2OS t-rex cells were seeded on 96-well plates (5,000 cells per well) in 100 ml DMEM (10% FBS, 1xPenStrep). Twenty four hours after transfection (day 0), medium was replaced with 100 ml DMEM (10% FBS, 1xPenStrep). Cell number was measured daily by incubating cells for 1.5 h with DMEM containing 1/10 diluted WST-8 reagent at 37 C (Cell Counting Kit-8, Sigma-Aldrich) and absorbance was read at 450 nm using an EnSpire Reader (Perkin Elmer).

Fluorescent microscopy. U2OS t-rex Rev-VNP (tetO-MYC WT/V393D or siRNA transfected) cells were seeded (10,000 cells per well) on ImageLock 96-well plates (Essen Bioscience) in 100 ml DMEM (10% FBS, 1xPenStrep). Next day, after dexamethasone synchronization and PBS wash, cells were incubated in 100 ml PBS- or Doxycycline-containing DMEM (10% FBS, 1xPenStrep) and monitored with IncuCyte ZOOM reader (Essen Bioscience). To detect cells in G0G1 phase, U2OS overexpressing MYC alleles were stably transfected with FUCCI-Red construct and puromycin-resistant clones were selected. Then 7,000 cells were seeded on ImageLock 96-well plates (Essen Bioscience) and next day, after addition of doxycycline, cells were monitored with an IncuCyte ZOOM reader (Essen Bioscience) and red-uorescent objects were counted using in-built software. Total uorescence and number of uorescent objects were measured using in-built software.

FACS. After 3 days of doxycycline induction, trypsinized and PBS-washed cells (B106 cells) were xed in 70% ethanol for 1 h at 4 C. After centrifugation for5 min at 300g, cells were washed once with PBS and incubated for 10 min at room temperature in 1 ml propidium iodide-staining solution (PBS, 0.1% TritonX-100, 10 mg ml 1 freshly added propidium iodide, 100 mg ml 1 freshly added RNaseA).

Cellular DNA content was measured with a FACSCanto analyzer (BD).

Co-immunoprecipitation (Co-IP) and western blotting. For protein analysis, synchronized U2OS cells were lysed in ice-cold lysis buffer (25 mM Tris-HCl, pH 8.0, 150 mM NaCl, 0.5% Triton X100, 2 mM EDTA, 1 mM NaF and protease inhibitor cocktail (Roche, 04693159001)) for 10 min on ice and sonicated in ice water in an ultrasonic bath (Merck) for 10 min. Then lysates were pre-cleared from cell debris by centrifugation at 16,000g for 15 min at 4 C and the protein concentration in the supernatant was determined with Roti-Quant (Carl Roth). Samples (200 mg of total protein) were boiled with the appropriate amount of

4x Laemmli buffer (250 mM Tris-HCl pH 6.8, 6% SDS, 40% Glycerol, 0.04% Bromphenolblue and 20% Mercaptoethanol) for 3 min at 95 C and separated using 12% SDS-PAGE. After semi-dry transfer (PEQLAB) on nitrocellulose membranes (GE Healthcare), protein extracts were decorated with anti-BMAL1 (ref. 13; 1:750), anti-CLOCK (ref. 13; 1:500), anti-MYC (1:400, N-262, SantaCruz), anti-V5 (1:5,000, 46-0705, Life Technologies) and anti-FLAG (1:5,000, M2, Sigma-Aldrich) antibodies in 5% milk TBS at 4 C overnight. Next day, membrane was washed three times in TBS and incubated with the appropriate HRP-conjugated secondary antibody in 5% milk TBS for 1 h at room temperature. After 4 TBS washes, the membrane was exposed to X-ray lms (Super RX, FUJIFILM) and developed using an AGFA automatic processor developer. For co-IP, lysates from transfected HEK293T cells were prepared as described above. Lysates (500 mg total protein) were incubated with anti-V5 (0.5 ml) antibodies or 40 ml of PBS-washed anti-FLAG M2 Afnity Gel (Sigma-Aldrich) agitating at 4 C overnight. Next day, 40 ml of PBS-washed protein A sepharose beads (GE Healthcare) were added to bind anti-V5 immune complexes and incubated for 2 h. Then beads were washed three times with PBS and protein was eluted by boiling in 4x Laemmli buffer. Precipitated proteins were analysed by western blotting as described above. Uncropped blots are shown in Supplementary Fig. 7.

References

1. Dunlap, J. C. Molecular bases for circadian clocks. Cell 96, 271290 (1999).2. Ukai, H. & Ueda, H. R. Systems biology of mammalian circadian clocks. Annu. Rev. Physiol. 72, 579603 (2010).

3. Sahar, S. & Sassone-Corsi, P. Metabolism and cancer: the circadian clock connection. Nat. Rev. Cancer 9, 886896 (2009).

4. Kelleher, F. C., Rao, A. & Maguire, A. Circadian molecular clocks and cancer. Cancer Lett. 342, 918 (2014).

5. Filipski, E. et al. Effects of chronic jet lag on tumor progression in mice. Cancer Res. 64, 78797885 (2004).

6. Filipski, E. et al. Host circadian clock as a control point in tumor progression.J. Natl Cancer Inst. 94, 690697 (2002).7. Cadenas, C. et al. Loss of circadian clock gene expression is associated with tumor progression in breast cancer. Cell Cycle 13, 32823291 (2014).

8. Fu, L. & Kettner, N. M. The circadian clock in cancer development and therapy. Prog. Mol. Biol. Transl. Sci. 119, 221282 (2013).

9. DeMarzo, A. M., Nelson, W. G., Isaacs, W. B. & Epstein, J. I. Pathological and molecular aspects of prostate cancer. Lancet 361, 955964 (2003).

10. Adhikary, S. & Eilers, M. Transcriptional regulation and transformation by Myc proteins. Nat. Rev. Mol. Cell Biol. 6, 635645 (2005).

11. Larsson, L. G. & Henriksson, M. A. The Yin and Yang functions of the Myc oncoprotein in cancer development and as targets for therapy. Exp. Cell Res. 316, 14291437 (2010).

12. Altman, B. J. et al. MYC disrupts the circadian clock and metabolism in cancer cells. Cell Metab. 22, 10091019 (2015).

13. Hoffmann, J. et al. Non-circadian expression masking clock-driven weak transcription rhythms in U2OS cells. PLoS ONE 9, e102238 (2014).

14. Walz, S. et al. Activation and repression by oncogenic MYC shape tumourspecic gene expression proles. Nature 511, 483487 (2014).

15. Hurlin, P. J., Queva, C. & Eisenman, R. N. Mnt, a novel Max-interacting protein is coexpressed with Myc in proliferating cells and mediates repression at Myc binding sites. Genes Dev. 11, 4458 (1997).

16. Nagoshi, E. et al. Circadian gene expression in individual broblasts: cell-autonomous and self-sustained oscillators pass time to daughter cells. Cell 119, 693705 (2004).

17. Baggs, J. E. et al. Network features of the mammalian circadian clock. PLoS Biol. 7, e52 (2009).

18. Staller, P. et al. Repression of p15INK4b expression by Myc through association with Miz-1. Nat. Cell Biol. 3, 392399 (2001).

19. Wu, S. et al. Myc represses differentiation-induced p21CIP1 expression via Miz-1-dependent interaction with the p21 core promoter. Oncogene 22, 351360 (2003).

20. Wolf, E. et al. Miz1 is required to maintain autophagic ux. Nature Commun. 4, 2535 (2013).

21. Herold, S. et al. Negative regulation of the mammalian UV response by Myc through association with Miz-1. Mol. Cell 10, 509521 (2002).22. Dang, C. V. MYC on the path to cancer. Cell 149, 2235 (2012).23. van Riggelen, J. et al. The interaction between Myc and Miz1 is required to antagonize TGFbeta-dependent autocrine signaling during lymphoma formation and maintenance. Genes Dev. 24, 12811294 (2010).

24. Cho, R. J. et al. Transcriptional regulation and function during the human cell cycle. Nat. Genet. 27, 4854 (2001).

25. Grant, G. D. et al. Identication of cell cycle-regulated genes periodically expressed in U2OS cells and their regulation by FOXM1 and E2F transcription factors. Mol. Biol. Cell 24, 36343650 (2013).

26. Storch, K. F. et al. Extensive and divergent circadian gene expression in liver and heart. Nature 417, 7883 (2002).

27. Panda, S. et al. Coordinated transcription of key pathways in the mouse by the circadian clock. Cell 109, 307320 (2002).

NATURE COMMUNICATIONS | 7:11807 | DOI: 10.1038/ncomms11807 | http://www.nature.com/naturecommunications

Web End =www.nature.com/naturecommunications 9

ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/ncomms11807

28. Matsuo, T. et al. Control mechanism of the circadian clock for timing of cell division in vivo. Science 302, 255259 (2003).

29. Bieler, J. et al. Robust synchronization of coupled circadian and cell cycle oscillators in single mammalian cells. Mol. Syst. Biol. 10, 739 (2014).

30. Feillet, C. et al. Phase locking and multiple oscillating attractors for the coupled mammalian clock and cell cycle. Proc Natl Acad Sci USA 111, 98289833 (2014).

31. Hughes, M. E. et al. Harmonics of circadian gene transcription in mammals. PLoS Genet. 5, e1000442 (2009).

32. Wiese, K. E. et al. Repression of SRF target genes is critical for Myc-dependent apoptosis of epithelial cells. EMBO J. 34, 15541571 (2015).

33. Brown, S. A. et al. PERIOD1-associated proteins modulate the negative limb of the mammalian circadian oscillator. Science 308, 693696 (2005).

34. Vanselow, K. et al. Differential effects of PER2 phosphorylation: molecular basis for the human familial advanced sleep phase syndrome (FASPS). Genes Dev. 20, 26602672 (2006).

35. Sporl, F. et al. A circadian clock in HaCaT keratinocytes. J. Invest. Dermatol. 131, 338348 (2011).

36. Nelson, J. D., Denisenko, O. & Bomsztyk, K. Protocol for the fast chromatin immunoprecipitation (ChIP) method. Nat. Protoc. 1, 179185 (2006).

37. Richter, J. et al. Recurrent mutation of the ID3 gene in Burkitt lymphoma identied by integrated genome, exome and transcriptome sequencing. Nat. Genet. 44, 13161320 (2012).

38. Mortazavi, A., Williams, B. A., McCue, K., Schaeffer, L. & Wold, B. Mapping and quantifying mammalian transcriptomes by RNA-Seq. Nat. Methods 5, 621628 (2008).

Acknowledgements

We thank Steve Brown, Achim Kramer, Ueli Schibler and Elmar Wolf for plasmid constructs. We thank Monika Langlotz (ZMBH Flow Cytometry & FACS Core Facility).

This work was supported by funds of the Deutsche Forschungsgemeinschaft (SFB 1036 to MB). M.B. is a member of CellNetworks.

Author contributions

A.S. and B.R. performed the experiments. A.S., A.D. and M.B. planned and designed the experiments and wrote the manuscript. N.H., P.B., U.H.T. and M.S. performed bioinformatics analyses of expression data. M.S. and R.E. supervised bioinformatics analyses. The ICGC MMML-seq consortium contributed gene expression data for human lymphomas.

Additional information

Supplementary Information accompanies this paper at http://www.nature.com/naturecommunications

Web End =http://www.nature.com/ http://www.nature.com/naturecommunications

Web End =naturecommunications

Competing nancial interests: The authors declare no competing nancial interests.

Reprints and permission information is available online at http://npg.nature.com/reprintsandpermissions/

Web End =http://npg.nature.com/ http://npg.nature.com/reprintsandpermissions/

Web End =reprintsandpermissions/

How to cite this article: Shostak, A. et al. MYC/MIZ1-dependent gene repression inversely coordinates the circadian clock with cell cycle and proliferation. Nat. Commun. 7:11807 doi: 10.1038/ncomms11807 (2016).

This work is licensed under a Creative Commons Attribution 4.0 International License. The images or other third party material in this article are included in the articles Creative Commons license, unless indicated otherwise in the credit line; if the material is not included under the Creative Commons license, users will need to obtain permission from the license holder to reproduce the material. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/

Web End =http://creativecommons.org/licenses/by/4.0/

Members of the ICGC MMML-Seq Project

Chris Lawerenz2, Peter Lichter2, Bernhard Radlwimmer2, Jrgen Eils2, Benedikt Brors2, Sylwester Radomski2, Ingrid Scholz2, Gesine Richter4, Reiner Siebert4, Susanne Wagner4, Andrea Haake4, Julia Richter4,Sietse Aukema4, Ole Ammerpohl4, Christina Lopez4, Inga Nagel4, Inga Vater4, Rabea Wagner4,Christoph Borst4, Siegfried Haas5, Marius Rohde6, Birgit Burkhardt6,7, Jasmin Lisfeld7, Alexander Claviez8, Martin Dreyling9, Sonja Eberth10, Lorenz Trmper10, Dieter Kube10, Christina Stadler10, Hermann Einsele11, Norbert Frickhofen12, Martin-Leo Hansmann13, Dennis Karsch14, Michael Kneba14, Luisa Mantovani-Lfer15, Peter Staib16, Stephan Stilgenbauer17, German Ott18, Ralf Kppers19, Marc Weniger19, Michael Hummel20, Dido Lenze20, Monika Szczepanowski21, Wolfram Klapper21, Ulrike Kostezka22, Peter Mller23, Andreas Rosenwald24, Ellen Leich24, Jordan Pischimariov24, Vera Binder25, Arndt Borkhardt25, Kebria Hezaveh25, Jessica Hoell25, Philip Rosenstiel26, Markus Schilhabel26, Stefan Schreiber27, Stephan H. Bernhart28-30, Gero Doose2830, Steve Hoffmann2830, Helene Kretzmer2830, David Langenberger28-30, Hans Binder29, Lydia Hopp29, Markus Kreuz31, Markus Loefer31, Maciej Rosolowski31, Jan Korbel32, Stefanie Sungalee32, Peter F. Stadler2830,3335, Thorsten Zenz36

4Institute of Human Genetics, University Hospital Schleswig-Holstein Campus Kiel/ Christian-Albrechts University Kiel, D-24105 Kiel, Germany;

5Friedrich-Ebert Hospital Neumnster, Clinics for Hematology, Oncology and Nephrology, D-24534 Neumnster, Germany; 6Department of Pediatric Hematology and Oncology, University Hospital Mnster, D-48149 Mnster, Germany; 7Department of Pediatric Hematology and Oncology University Hospital Giessen, D-35043 Giessen, Germany; 8Department of Pediatrics, University Hospital Schleswig-Holstein, Campus Kiel, D-24105 Kiel, Germany;

9Department of Medicine III - Campus Grosshadern, University Hospital Munich, D-81377 Munich, Germany; 10Department of Hematology and Oncology, Georg-August-University of Gttingen, D-37075 Gttingen, Germany; 11University Hospital Wrzburg, Department of Medicine and Poliklinik II, University of Wrzburg, D-97080 Wrzburg, Germany; 12Department of Medicine III, Hematology and Oncology, Dr. Horst-Schmidt-Kliniken of Wiesbaden, D-65199 Wiesbaden, Germany; 13Senckenberg Institute of Pathology, University of Frankfurt Medical School, D-60590 Frankfurt am Main, Germany; 14Department of Internal Medicine II: Hematology and Oncology, University Medical Centre, Campus Kiel, D-24105 Kiel, Germany; 155Hospital of Internal Medicine II, Hematology and Oncology, St-Georg Hospital Leipzig, D-04129 Leipzig, Germany; 16Univesity Hospital Aachen, St.-Antonius Hospital, Department of Oncology, Hematology and stem cell transplantation, University of Aachen, D-52074 Aachen, Germany; 17Department of Internal Medicine III, University of Ulm, D-89081 Ulm, Germany; 18Robert-Bosch Hospital Stuttgart, Department of Pathology, D-70376 Stuttgart, Germany; 19Institute of Cell Biology (Cancer Research), University of Duisburg-Essen, D-45122 Essen, Germany; 20Institute of Pathology, Charit - University Medicine Berlin, D-10117 Berlin, Germany; 21Hematopathology Section, University Hospital Schleswig-Holstein Campus Kiel/Christian-Albrechts University Kiel, D-24105 Kiel, Germany;

10 NATURE COMMUNICATIONS | 7:11807 | DOI: 10.1038/ncomms11807 | http://www.nature.com/naturecommunications

Web End =www.nature.com/naturecommunications

NATURE COMMUNICATIONS | DOI: 10.1038/ncomms11807 ARTICLE

22Comprehensive Cancer Center Ulm (CCCU), University Hospital Ulm, D-89081 Ulm, Germany; 23Institute of Pathology, Medical Faculty of the Ulm University, D-89081 Ulm, Germany; 24Institute of Pathology, University of Wrzburg, D-97080 Wrzburg, Germany; 25Department of Pediatric Oncology, Hematology and Clinical Immunology, Heinrich-Heine-University, D-40225 Dsseldorf, Germany; 26Institute of Clinical Molecular Biology, University Hospital Schleswig-Holstein Campus Kiel/Christian-Albrechts University Kiel, D-24105 Kiel, Germany; 27Department of General Internal Medicine, University Hospital Schleswig-Holstein Campus Kiel/Christian-Albrechts University Kiel, D-24105 Kiel, Germany; 28Transcriptome Bioinformatics Group, LIFE Research Center for Civilization Diseases, D-04103 Leipzig, Germany; 29Interdisciplinary Center for Bioinformatics, University of Leipzig, D-04107 Leipzig, Germany; 30Bioinformatics Group, Department of Computer Science, University of Leipzig, D-04103 Leipzig, Germany; 31Institute for Medical Informatics Statistics and Epidemiology, D-04107 Leipzig, Germany; 32EMBL Heidelberg, Genome Biology, D-69117 Heidelberg, Germany; 33RNomics Group, Fraunhofer Institute for Cell Therapy and Immunology IZI, D-04103 Leipzig, Germany; 34Santa Fe Institute, Santa Fe, NM 87501, United States of America;

35Max-Planck-Institute for Mathematics in Sciences, D-04103 Leipzig, Germany; 36Department of Medicine V, University of Heidelberg, D-69120 Heidelberg, Germany.

NATURE COMMUNICATIONS | 7:11807 | DOI: 10.1038/ncomms11807 | http://www.nature.com/naturecommunications

Web End =www.nature.com/naturecommunications 11