Báo cáo khoa học: Nuclear import of mPER3 in Xenopus oocytes and HeLa cells requires complex formation with mPER1 pdf

Mammalian PER proteins have also been found to become phosphorylated; mPER1, mPER2 and mPER3 are subjected to rhythmical phosphorylation mediated Keywords circadian rhythm; mCRY, mPER; n

cells requires complex formation with mPER1

Susanne Loop and Tomas Pieler

Abteilung Entwicklungsbiochemie, Zentrum fu¨r Biochemie und Molekulare Zellbiologie, Georg-August Universita¨t, Go¨ttingen, Germany

The genetic control of circadian rhythmicity was first

analysed in Drosophila A central autoregulatory

feed-back loop that involves different transcriptional

regula-tors was uncovered The bHLH transcription facregula-tors

CLOCK (CLK) and CYCLE (CYC) drive expression

of the period (per) and timeless (tim) genes Conversely,

Period and Timeless proteins (PER and TIM) inhibit

CLK⁄ CYC-mediated transcription of their own genes,

resulting in a gradual loss of PER and TIM proteins

At a critically reduced level of PER and TIM protein

activity, CLK⁄ CYC repression is relieved and per ⁄ tim

gene expression returns [1–5]

A similar mechanism seems to operate in

verte-brates In mammals, CLOCK–BMAL1 heterodimers

activate transcription of Period (mPer) and

Crypto-chrome (mCry) genes mPER and mCRY proteins act

as negative regulators of their own expression by

directly interacting with and thereby inhibiting

CLOCK–BMAL1 [6,7] Gene duplications have

gener-ated three different mPER proteins (mPER1, mPER2

and mPER3) and two different mCRY proteins

(mCRY1 and mCRY2) Functional diversity among the individual members of each of these clock protein subfamilies has been reported [8–15]

Post-translational control constitutes a further important level of regulation in both vertebrate and invertebrate systems In Drosophila, phoshorylation of both PER and TIM affects stability and⁄ or nuclear transport [16–21] Phosphorylated forms of the two proteins are targeted for degradation by the ubiquitin– proteasome pathway [22–24] It has also been pro-posed that PER⁄ TIM phosphorylation may promote nuclear transfer, but more recent studies argue in favour of phosphorylation positively regulating their transcriptional repressor activity [25] Conversely, the regulated rhythmic dephosphorylation of PER by pro-tein phosphatase 2A stabilizes PER, thereby contribu-ting to the rhythmicity of PER protein concentrations [26]

Mammalian PER proteins have also been found to become phosphorylated; mPER1, mPER2 and mPER3 are subjected to rhythmical phosphorylation mediated

Keywords

circadian rhythm; mCRY, mPER; nuclear

import; Xenopus oocytes

Correspondence

T Pieler, Abteilung Entwicklungsbiochemie,

Zentrum fu¨r Biochemie und Molekulare

Zellbiologie, Georg-August Universita¨t,

Justus von Liebig Weg 11, D-37077

Go¨ttingen, Germany

Fax: +49 551 3914614

Tel: +49 551 395683

E-mail: tpieler@gwdg.de

(Received 14 March 2005, revised 27 May

2005, accepted 31 May 2005)

doi:10.1111/j.1742-4658.2005.04798.x

Several transcription factors with the function of setting the biological clock in vertebrates have been described A detailed understanding of their nucleocytolasmic transport properties may uncover novel aspects of the regulation of the circadian rhythm This assumption led us to perform a systematic analysis of the nuclear import characteristics of the different murine PER and CRY proteins, using Xenopus oocytes and HeLa cells as experimental systems Our major finding is that nuclear import of mPER3 requires complex formation with mPER1 We further show that the nuclear localization signal (NLS) function of mPER1 and not activation of a masked NLS in mPER3 is critical for the import of the mPER1–mPER3 complex Finally, and as previously described in other cell systems, nuclear import of mPER proteins in Xenopus oocytes correlates positively with their phosphorylation

Abbreviations

CK, casein kinase; NLS, nuclear localization signal.

by casein kinases (CKIe and CKId) [27–30] The

phos-phorylation status of murine PER proteins, similar to

that reported for Drosophila PER, influences stability

and nuclear transport; phosphorylated forms of

mPER1 and mPER3 are rapidly degraded [31,32]

Furthermore, mPER1 mutant mice have been used to

demonstrate that mPER1 is required for

phosphoryla-tion and nuclear transfer of mPER3 [33]

Phosphoryla-tion of mPER1 itself correlates with nuclear transport

[34] Earlier studies had already indicated that nuclear

translocation of mPER3 is promoted by mPER1 in

NIH3T3 cells [14]

Other studies, using different cell systems, had come

to additional and sometimes apparently contradictory

conclusions Yagita et al [35], using COS7 cells,

repor-ted that mPER3 by itself is predominantly

cytoplas-mic, and nuclear accumulation is obtained by serum

shock-induced formation of mPER1⁄ 3 or mPER2 ⁄ 3

heterodimers Furthermore, Vielhaber et al [36] had

observed that mPER1 is predominantly nuclear,

whereas mPER2 is predominantly cytoplasmic in

HEK293 cells; CKIe-mediated phosphorylation of

mPER1 was reported to lead to masking of the nuclear

localization signal (NLS) and coexpression of mPER1

with mPER2 and cytoplasmic localization of the

heterodimer Finally, Miyazaki et al [37], using COS1

cells, had observed that mammalian PER2 has a

posit-ive regulatory function with respect to the nuclear

import of mCRY1

What all these studies have in common is the idea that dimerization of the different mammalian PER and CRY proteins modulates their nucleocytoplasmic dis-tribution and thereby probably also their function as transcriptional repressors In the work presented here,

we systematically analyzed nuclear import of the dif-ferent murine PER and CRY proteins, either individu-ally or in all possible heterodimeric combinations, primarily using Xenopus oocytes as an experimental system We found that interaction with mPER1 is required for the nuclear import of mPER3, and we observed a positive correlation between nuclear import

of mPER proteins and their phosphorylation

Results

Positive correlation between phosphorylation and nuclear import of mPER proteins in Xenopus oocytes

The different individual murine PER and CRY pro-teins were generated by in vitro translation and injected into the cytoplasm of Xenopus oocytes The kinetics of nuclear import were analysed after manual separation

of cytoplasmic and nuclear fractions by gel electro-phoresis (Fig 1A) The data obtained reveal that, whereas the mCRY proteins, as well as mPER1 and mPER2, are readily imported into the nucleus of Xeno-pusoocytes, mPER3 is not We also observed reduced

A

B

Fig 1 Nuclear import of murine PER and CRY proteins in Xenopus oocytes mPER1 and mPER2, but not mPER3, are phosphorylated and imported into the nucleus of Xenopus oocytes (A) 35 S-labelled mPER1, mPER2, mPER3 and derived protein fragments fused to six copies

of the myc tag in tandem repeat were translated in vitro and injected into the cytoplasm of Xenopus oocytes The nucleus and cytoplasm were separated manually at the time points indicated Proteins were immunoprecipitated from 10 pooled nuclear and cytoplasmic fractions using the myc antibody and analysed by SDS ⁄ PAGE and phosphorimaging To test for phosphorylation, immunoprecipitated nuclear or cyto-plasmic fractions were incubated with lambda protein phosphatase (kPPase) before gel electrophoresis (B) mCRY1 and mCRY2 are impor-ted into the nucleus of Xenopus oocytes.

electrophoretic mobility of mPER1 and mPER2, but

not of mPER3, which increases with the time of

incu-bation after microinjection It also seems that the

relative amount of the phosphorylated forms of the

proteins is higher in the nucleus than in the cytoplasm

Reduced electrophoretic mobility suggests chemical

modification events, such as phosphorylation

Phos-phatase treatment of cytoplasmic and nuclear protein

fractions isolated from microinjected oocytes equalizes

the electrophoretic mobility of all samples tested,

revealing that mPER1 and mPER2 are indeed

phos-phorylated after injection into Xenopus oocytes Thus,

we found a positive correlation between

phosphoryla-tion and nuclear import for mPER1 and mPER2,

whereas mPER3, which is not imported into the

nuc-leus, is also not phosphorylated On the other hand,

there is no evidence for phosphorylation of mCRY1

and mCRY2, which are readily imported into the

nuc-leus of injected oocytes (Fig 1B)

The absence of nuclear import of mPER3 injected into Xenopus oocytes suggests that the protein is devoid of a nuclear import signal that is functional in this experimental system To address this question, all three murine PER proteins were broken down into four fragments, and each one tested for nuclear import activity in Xenopus oocytes (Fig 2) In agree-ment with earlier NLS-mapping experiagree-ments in other experimental systems [31,35–37], the corresponding region (fragment 3) of all three PER proteins har-bours a functional NLS Mutation of the putative NLS in mPER3 abrogates import activity (data not shown) In extension of previous studies, we further detected a novel, additional NLS located in the C-ter-minal portion (fragment 4) of mPER1 within the 186 C-terminal amino acids (Fig 3, fragment 4b) We also noted faint nuclear signals for the corresponding C-terminal fragments derived from mPER2 and mPER3 (Fig 2A) However, nuclear import rates

A

B

Fig 2 mPER1 contains an additional NLS in its C-terminal domain (A) Mapping of NLS function in murine Per proteins Fragments corres-ponding to different portions of mPER1, mPER2 and mPER3 (as indicated) were assayed for nuclear import in Xenopus oocytes MPER2 Frag2 was rapidly degraded in Xenopus oocytes mPER1: myc Frag 1, aa 1–323; myc Frag 2, aa 324–645; myc Frag 3, aa 646–972; myc Frag

4, aa 973–1291 mPER2: myc Frag 1, 1–314; myc Frag 2, aa 314–628; myc Frag 3, 629–942; myc Frag 4, aa 943–1257 mPER3: myc Frag 1,

aa 1–280; myc Frag 2, 281–559; myc Frag 3, 560–835; myc Frag 4, 836–1113 (B) Schematic representation of the fragments used for map-ping experiments and percentage of nuclear import in multiple independent experiments The grey boxes define the location of the NLSs in the Per proteins (in bold the newly identified NLS2 in mPER1).

below 10% (Fig 2B) are considered nonsignificant A

primary sequence comparison of the three murine

PER proteins revealed a high degree of structural

diversity in the C-terminal domain (data not shown),

correlating with functional diversity with respect to

NLS activities Mutation or deletion of one of the

two NLSs in mPER1 led to reduced nuclear import

A complete block occurred only after mutation⁄

dele-tion of both NLSs (Fig 3, myc- mPER1mutNLSDC);

phosphorylation was not affected in these mutants

(data not shown)

Alternative explanations exist for the observed

absence of mPER3 nuclear import when injected by

itself; either it is rapidly degraded in the nucleus or

rapid export prevents its nuclear accumulation

How-ever, in a separate study on the nuclear export of clock

proteins in Xenopus oocytes [38], we observed that,

after nuclear injection of mPER3, the protein is only

slowly exported and there is no indication of protein

degradation in the nucleus

Thus, in summary, microinjection of individual

iso-lated murine PER proteins reveals that mPER1 and

mPER2 become phosphorylated and are imported into

the nucleus of Xenopus oocytes In contrast, mPER3 is

not phosphorylated and not transferred to the nucleus,

even though it contains an NLS that is functional in

this system Furthermore, deletion analysis uncovered

a novel NLS (NLS2) that is specific to the C-terminal region of mPER1

Complex formation with mPER1 promotes nuclear import of mPER3 in Xenopus oocytes

As heterodimerization of clock proteins is known to modulate nuclear import activity, we tested whether complex formation with either mPER1 or mPER2 would enable transfer of mPER3 into the nucleus of Xenopus oocytes For this purpose, mPER dimers were formed in vitro (Fig 4A); we found that cotranslation

of different combinations of mPER proteins allowed heterodimerization, whereas coincubation after in vitro translation did not In good agreement with earlier studies [35,39], we also found that the entire PAS domain in mPER1 was required for complex forma-tion with mPER3 (data not shown), and the NLS-defi-cient mPER1 mutant was not impaired with respect to its ability to interact with mPER3 (Fig 4A)

Microinjection of complexes formed with different combinations of mPER proteins into the cytoplasm of Xenopus oocytes revealed that, whereas mPER3 by itself (Fig 1A) or in complex with mPER2 was not imported, it was readily transferred to the nucleus in complex with mPER1 (Fig 4B) As expected, a com-plex of mPER1 and mPER2 was also imported Thus,

A

B

Fig 3 Mutation of the NLS function in

mPER1 blocks nuclear import activity (A)

Different mutants of mPER1 (as indicated)

were assayed for nuclear import activity in

Xenopus oocytes To mutate mPER1-NLS1,

three of the basic amino acids were

chan-ged to alanine (RRHHCRSKAKRSR) In

mPER1DC and mPER1mutNLS1DC, the 186

C-terminal amino-acid sequence containing

NLS2 was deleted myc mPER1 Frag 4a,

aa 973–1104; myc mPER1 Frag 4b,

1105–1291; myc mPER1mutNLS1, aa

1–1291; myc mPER1DC, aa 1–1104; myc

mPER1mutNLS1DC, aa 1–1291 (B)

Percent-age of nuclear import of multiple

independ-ent experimindepend-ents.

mPER1 seems to serve as an adaptor for the nuclear

import of mPER3 in Xenopus oocytes

As both mPER1 and mPER3 contain functional

NLSs (as described above), we tested whether complex

formation with mPER1 would unmask the NLS activity

in full-length mPER3 We constructed a mutant version

of mPER1 that had lost both of its two NLSs but

retained its ability to form a heterodimer with mPER3 (mPER1mutNLS1DC; Figs 3 and 4A) In complex with this mutant mPER1 variant, mPER3 was no longer transferred to the nucleus (Fig 4C) Conversely, upon mutation of the NLS in mPER3, the mPER1⁄ mPER3mutNLS heterodimer was still imported into the nucleus of Xenopus oocytes (Fig 4C), suggesting that it

A

Fig 4 mPER3 is imported into the nucleus of Xenopus oocytes in complex with mPER1 (A) Homodimer and heterodimer formation of mPER proteins Flag-tagged mPER3 was cotranslated in vitro with myc-tagged versions of full-length mPER1, mPER2, mPER3 and mPER1mutNLS1DC Complex formation was detected by coimmunoprecipitation using a flag antibody (bottom panel) As a negative control, myc tagged period proteins were translated without flag mPER3 and immunprecipitated by using the flag antibody (left hand panel) 50% of the input was loaded on the SDS ⁄ polyacrylamide gel (B) Complexes formed by cotranslation of different combinations of myc-tagged mPER1, mPER2 and mPER3 (as indicated) were injected into the cytoplasm of Xenopus oocytes and assayed for nuclear import after 3 and

6 h incubation at 18
C as described in Fig 1 (C) The NLS function in mPER1 is required for mPER3 import The heterodimer of myc mPER3 and flag mPER1mutNLS1DC was injected into the cytoplasm of Xenopus oocytes; nuclear and cytoplasmic fractions were immuno-precipitated by using the myc and flag antibodies at the time points indicated Myc-tagged, cotranslated mPER1 and mPER3mutNLS were analysed for nuclear import All proteins were treated with lambda protein phosphatase before electrophoresis.

Fig 5 The NLS functions of mPER proteins are also active in HeLa cells (A) Schematic representation of mPER proteins and derived fragments used for transient transfection into HeLa cells and their nucleocytoplasmic distribution (B) HeLa cells were transfected with the indicated myc-tagged mPER proteins The intracellular localization of these proteins was detected by immunofluorescence staining using Cy3-coupled myc antibodies (red) The nuclei were visualized by DAPI DNA staining (blue) (C) Quantitative analysis The subcellular localiza-tion of the different protein constructs was categorized as nuclear (N), nuclear and cytoplasmic (N ⁄ C), or cytoplasmic (C) For each construct, 50–100 transfected cells were analysed.

B

C

is the NLS activity in mPER1, and not unmasking of

the NLS in mPER3, that is responsible for the nuclear

transfer of the mPER1–mPER3 complex

Nuclear import of mPER3 in HeLa cells also

requires complex formation with mPER1

We further investigated whether the above import

characteristics of murine PER proteins reflect specific

features of nucleocytoplasmic transport in Xenopus

oocytes HeLa cells were transiently transfected with

the same set of mPER protein fragments as used in

the oocyte microinjection experiments We found that

the main effects, i.e the lack of nuclear import of

mPER3 and the presence of an additional NLS at the

C-terminus of mPER1, can be reproduced in these cells

(Fig 5) In addition, we also observed weak nuclear

import activity for the C-terminal fragment of mPER2

(Fig 5, mPER2 Frag 4)

Next, we analysed whether, similar to the situation

with Xenopus oocytes, complex formation with mPER1

is sufficient for nuclear import of mPER3 in HeLa

cells mPER proteins alone, or specific combinations

of mPER3 with mPER1 or mPER1mutNLS1DC, were

used in the transient transfection assay (Fig 6)

Indeed, in combination with mPER1, but not with

mPER2, mPER3 was mostly nuclear; analysis of a

combination of mPER3 with mPER1mutNLS1DC

revealed that, again as in the oocyte system, it is the

NLS function of mPER1 that is required for the

nuc-lear import of mPER3 in the heterodimeric complex

with mPER1 Thus, the requirement of complex

for-mation with mPER1 for the nuclear import of mPER3

appears to be a general phenomenon that is not

restricted to the Xenopus oocyte system

Discussion

Analysis of the nucleocytoplasmic transport activities

of murine PER and CRY proteins in Xenopus oocytes

and HeLa cells reveals that mPER1 serves as a nuclear

import adaptor for mPER3, even though mPER3

con-tains a functional NLS that appears to be masked in

the full-length protein We also mapped a novel NLS

to the C-terminus of mPER1 Nuclear import of the

mPER1–mPER3 complex requires a functional NLS in

mPER1, and the silent NLS in mPER3 is not necessary

Finally, nuclear import of mPER1 and mPER2

corre-lates with their phosphorylation in Xenopus oocytes

A systematic fragmentation analysis of the three

dif-ferent murine PER proteins produced two main

obser-vations First, mPER1 contains a second NLS at its

extreme C-terminus in addition to the one that had

been described previously [36], which is functional in both Xenopus oocytes and HeLa cells Secondly, mPER3 contains a silent NLS that is repressed in the context of the full-length protein The corresponding protein fragment contains a basic stretch of amino acids that is conserved in all three murine PER proteins Previous studies with COS7 cells also found cytoplasmic retention of mPER3 which could be relieved by cotransfection of CKIe [31] The molecular mechanism responsible for the masking of the NLS in mPER3 remains to be elucidated The NLS in mPER3 may be masked by intramolecular protein folding or

by interaction with an unknown inhibitory factor With respect to the elucidation of the mechanism that eventually relieves the cytoplasmic sequestration

of mPER3, previous studies used different cell lines and produced partially contradictory observations Our finding that the nuclear import of mPER3 is strongly enhanced in Xenopus oocytes and in HeLa cells by the presence of mPER1 is consistent with results obtained in COS7 and NIH3T3 cells [14,35] In further support of such a scenario, mPER3 has been reported to always be cytoplasmic in the livers of mPER1-deficient mice [33] However, Vielhaber et al [36] reported that coexpression of mPER1 with mPER2 results in cytoplasmic localization of the het-erodimer in HEK293 cells This result is inconsistent with our observations in microinjected oocytes and transiently transfected HeLa cells We cannot exclude the possibility that this apparent contradiction is a result of the use of different experimental systems Several independent studies also describe a positive correlation between mPER3 phosphorylation and nuclear accumulation [29,31,33] In Xenopus oocytes, cytoplasmic mPER3 was not found to be phosphoryl-ated, whereas nuclear import of mPER1 and mPER2 correlated with protein phosphorylation As mPER3 was also shown to require mPER1 for stable inter-action with CKIe and phosphorylation [33], there may

be a direct functional link between phoshorylation and activation of the ‘silent’ NLS in mPER3 However, Vielhaber et al [36] proposed that CKIe-mediated phosphorylation of mPER1 leads to NLS masking in HEK293 cells Again, this apparent contradiction may

be due to the differences in the experimental systems used

Experimental procedures

Plasmids For in vitro translation, mPer and mCry cDNAs were sub-cloned into the pCSMT vector containing six myc epitopes

[40], or into the pCSflag vector, in which the myc tag was

replaced by a double-stranded oligonucleotide sequence

containing a kozak element and the flag epitope (5¢-GATC

GCCGCCATGGACTACAAGGACGAGGATGACAA-3¢) The mPER2 cDNA was subcloned into the NcoI restriction site of pCSMT; the resulting construct possesses five copies

A

B

Fig 6 Nuclear import of clock proteins in

HeLa cells (A) The cells were transiently

transfected with myc-tagged and

flag-tagged proteins as indicated The

intracellu-lar localization of these proteins was

detec-ted by immunofluorescence staining using

myc-Cy3 (red) or flag-fluorescein

isothio-cyanate (FITC) (green) antibodies The nuclei

were visualized by DAPI DNA staining

(blue) (B) Quantitative analysis of the

nucleocytoplasmic distribution of mPER3

cotransfected with with other mPER

variants, as indicated (see also the legend to

Fig 5C).

of the myc epitope All mPER1 fragments were amplified

by PCR with 5¢ primers containing the EcoRI restriction

site and 3¢ primers containing the StuI restriction site All

mPER2 fragments were amplified by PCR with 5¢ primers

containing the NcoI restriction site and 3¢ primers

contain-ing the XhoI restriction site MPER3 Frag1 was amplified

by PCR with 5¢ primers containing the StuI restriction site

and 3¢ primers containing the XbaI restriction site mPER3

Frag2 and mPER3 Frag4 were amplified by PCR with 5¢

primers containing the EcoRI restriction site and 3¢ primers

containing the StuI restriction site MPER3 Frag3 was

amplified by PCR with 5¢ primers containing the EcoRI

restriction site and 3¢ primers containing the XbaI

restric-tion site The mPER1 mutants used were constructed by

using the Quick Change site-directed mutagenesis kit

(Strat-agene, La Jolla, CA, USA) using the user’s protocol

provi-ded by the manufacturer

Protein expression

Radiolabelled proteins were expressed as fusions with the

myc or flag epitope in a coupled transcription⁄ translation

(TNT) system (Promega, Madison, WI, USA) in the

pres-ence of 20 lCi [35S]methionine (Amersham, Little Chalfont,

Bucks, UK) The in vitro translated proteins products were

analysed by SDS⁄ PAGE and phosphorimaging (Molecular

Dynamics, Sunnyvale, CA, USA)

Coimmunoprecipitation experiments

For coimmunoprecipitation experiments, cDNAs were

mixed and in vitro cotranslated in the coupled TNT

sys-tem (Promega) The samples were incubated for 120 min

at 30
C, and 2 lL each sample added to protein

G–Seph-arose–myc–antibody pellets The coimmuoprecipitation

was performed in a final volume of NET-2 [50 mm

Tris⁄ HCl, pH 7.4, 150 mm NaCl, 0.05% (v ⁄ v) Nonidet

P40] for 1 h at 4
C After being washed three times with

NET-2, proteins were analysed by SDS⁄ PAGE and

phos-phorimaging

Microinjection into Xenopus laevis oocytes

Oocytes were prepared for microinjection as described

previously [41] All measures were taken to minimise pain

and discomfort of the frogs in accord with the German

regulations on experimental use of animals About 15 nL

protein injection mix was microinjected into the cytoplasm

of oocytes To determine the nucleocytoplasmic

distribu-tion, the nucleus and cytoplasm were manually separated

after different time intervals Proteins fused to the myc

epitope were purified from pooled nuclear and

cytoplas-mic fractions by immunoprecipitation as described by

Rudt & Pieler [42] The following antibodies were used:

mouse anti-myc (9E10; Sigma, St Louis, MO, USA) and mouse anti-flagM2 (Sigma)

Phosphatase treatment After immunoprecipitation, immunopellets were resuspended

in phosphatase buffer supplemented with 2 mm MnCl2and incubated with 200 U lambda protein phosphatase (New England Biolabs, Beverly, MA, USA) for 30 min at 30
C The addition of SDS⁄ PAGE sample buffer stopped the reac-tion

Cell culture and transfection Hela cells were cultured in Eagle’s minimal essential med-ium supplemented with 10% (v⁄ v) fetal bovine serum (Biochrom, Cambridge, UK) Approximately 3· 105

cells per well were plated in a six-well dish one day before trans-fection Plasmid (4 lg) was transfected with Lipofectamine

2000 (Invitrogen, San Diego, CA, USA) using the user’s protocol provided by the manufacturer

Immunocytochemistry The cells were grown on coverslips and fixed with 3% (v⁄ v) paraformaldehyde in NaCl⁄ Pi at room temperature for

15 min After treatment with 0.5% (v⁄ v) Triton X-100 in NaCl⁄ Pi, nonspecific staining was blocked with 3% (w⁄ v) BSA in NaCl⁄ Pi The immunostaining was performed with the myc-Cy3 or flag-FITC (Sigma) The cells were embed-ded with Vectashield containing 4’,6-diamidino-2-phenyl-indole (DAPI; Linaris, Bettingen, Germany)

Acknowledgements

This work was supported by a grant from the Deutsche Forschungsgemeinschaft (SFB 523) to T.P We thank

Dr Gregor Eichele and Dr Pablo Szendro for the mPer and mCry encoding plasmids, Dr Katja Koebernick for pCSflag, and Andreas Nolte for DNA sequencing

References

1 Rutila JE, Suri V, Le M, So WV, Rosbash M & Hall

JC (1998) CYCLE is a second bHLH-PAS clock protein essential for circadian rhythmicity and transcription of Drosophilaperiod and timeless Cell 93, 805–814

2 Lee C, Bae K & Edery I (1998) The Drosophila CLOCK protein undergoes daily rhythms in abundance, phos-phorylation, and interactions with the PER-TIM com-plex Neuron 21, 857–867

3 Lee C, Bae K & Edery I (1999) PER and TIM inhibit the DNA binding activity of a Drosophila CLOCK-CYC⁄

dBMAL1 heterodimer without disrupting formation

of the heterodimer: a basis for circadian transcription

Mol Cell Biol 19, 5316–5325

4 Darlington TK, Wager-Smith K, Ceriani MF, Staknis

D, Gekakis N, Steeves TD, Weitz CJ, Takahashi JS &

Kay SA (1998) Closing the circadian loop:

CLOCK-induced transcription of its own inhibitors per and tim

Science 280, 1599–1603

5 Allada R, White NE, So WV, Hall JC & Rosbash M

(1998) A mutant Drosophila homolog of mammalian

Clock disrupts circadian rhythms and transcription of

period and timeless Cell 93, 791–804

6 Bunger MK, Wilsbacher LD, Moran SM, Clendenin C,

Radcliffe LA, Hogenesch JB, Simon MC, Takahashi JS

& Bradfield CA (2000) Mop3 is an essential component

of the master circadian pacemaker in mammals Cell

103, 1009–1017

7 Gekakis N, Staknis D, Nguyen HB, Davis FC,

Wilsbacher LD, King DP, Takahashi JS & Weitz CJ

(1998) Role of the CLOCK protein in the mammalian

circadian mechanism Science 280, 1564–1569

8 Tei H, Okamura H, Shigeyoshi Y, Fukuhara C, Ozawa

R, Hirose M & Sakaki Y (1997) Circadian oscillation of

a mammalian homologue of the Drosophila period gene

Nature 389, 512–516

9 Zylka MJ, Shearman LP, Weaver DR & Reppert SM

(1998) Three period homologs in mammals: differential

light responses in the suprachiasmatic circadian clock

and oscillating transcripts outside of brain Neuron 20,

1103–1110

10 van der Horst GT, Muijtjens M, Kobayashi K, Takano

R, Kanno S, Takao M, de Wit J, Verkerk A, Eker AP,

van Leenen D, et al (1999) Mammalian Cry1 and Cry2

are essential for maintenance of circadian rhythms

Nature 398, 627–630

11 Sun ZS, Albrecht U, Zhuchenko O, Bailey J, Eichele G

& Lee CC (1997) RIGUI, a putative mammalian

ortho-log of the Drosophila period gene Cell 90, 1003–1011

12 Shearman LP, Sriram S, Weaver DR, Maywood ES,

Chaves I, Zheng B, Kume K, Lee CC, van der Horst

GT, Hastings MH & Reppert SM (2000) Interacting

molecular loops in the mammalian circadian clock

Science 288, 1013–1019

13 Shearman LP, Zylka MJ, Weaver DR, Kolakowski LF

Jr & Reppert SM (1997) Two period homologs:

circa-dian expression and photic regulation in the

suprachias-matic nuclei Neuron 19, 1261–1269

14 Kume K, Zylka MJ, Sriram S, Shearman LP, Weaver

DR, Jin X, Maywood ES, Hastings MH & Reppert SM

(1999) mCRY1 and mCRY2 are essential components

of the negative limb of the circadian clock feedback

loop Cell 98, 193–205

15 Albrecht U, Sun ZS, Eichele G & Lee CC (1997) A

differ-ential response of two putative mammalian circadian

reg-ulators, mper1 and mper2, to light Cell 91, 1055–1064

16 Akten B, Jauch E, Genova GK, Kim EY, Edery I, Raabe T & Jackson FR (2003) A role for CK2 in the Drosophilacircadian oscillator Nat Neurosci 6, 251–257

17 Bao S, Rihel J, Bjes E, Fan JY & Price JL (2001) The Drosophiladouble-timeS mutation delays the nuclear accumulation of period protein and affects the feedback regulation of period mRNA J Neurosci 21, 7117–7126

18 Lin JM, Kilman VL, Keegan K, Paddock B, Emery-Le

M, Rosbash M & Allada R (2002) A role for casein kinase 2alpha in the Drosophila circadian clock Nature

420, 816–820

19 Martinek S, Inonog S, Manoukian AS & Young MW (2001) A role for the segment polarity gene shaggy⁄

GSK-3 in the Drosophila circadian clock Cell 105, 769–779

20 Price JL, Blau J, Rothenfluh A, Abodeely M, Kloss B & Young MW (1998) double-time is a novel Drosophila clock gene that regulates PERIOD protein accumula-tion Cell 94, 83–95

21 Kloss B, Price JL, Saez L, Blau J, Rothenfluh A, Wesley CS & Young MW (1998) The Drosophila clock gene double-time encodes a protein closely related to human casein kinase Iepsilon Cell 94, 97–107

22 Naidoo N, Song W, Hunter-Ensor M & Sehgal A (1999) A role for the proteasome in the light response

of the timeless clock protein Science 285, 1737–1741

23 Grima B, Lamouroux A, Chelot E, Papin C, Limbourg-Bouchon B & Rouyer F (2002) The F-box protein slimb controls the levels of clock proteins period and timeless Nature 420, 178–182

24 Ko HW, Jiang J & Edery I (2002) Role for Slimb in the degradation of Drosophila Period protein phosphory-lated by Doubletime Nature 420, 673–678

25 Nawathean P & Rosbash M (2004) The doubletime and CKII kinases collaborate to potentiate Drosophila PER transcriptional repressor activity Mol Cell 13, 213–223

26 Sathyanarayanan S, Zheng X, Xiao R & Sehgal A (2004) Posttranslational regulation of Drosophila PER-IOD protein by protein phosphatase 2A Cell 116, 603– 615

27 Lowrey PL, Shimomura K, Antoch MP, Yamazaki S, Zemenides PD, Ralph MR, Menaker M & Takahashi

JS (2000) Positional syntenic cloning and functional characterization of the mammalian circadian mutation tau Science 288, 483–492

28 Keesler GA, Camacho F, Guo Y, Virshup D, Mondadori C & Yao Z (2000) Phosphorylation and destabilization of human period I clock protein by human casein kinase I epsilon Neuroreport 11, 951–955

29 Takano A, Shimizu K, Kani S, Buijs RM, Okada M & Nagai K (2000) Cloning and characterization of rat casein kinase 1epsilon FEBS Lett 477, 106–112

30 Toh KL, Jones CR, He Y, Eide EJ, Hinz WA, Virshup

DM, Ptacek LJ & Fu YH (2001) An hPer2 phosphory-lation site mutation in familial advanced sleep phase syndrome Science 291, 1040–1043