Constitutive Expression and Delayed LightResponse of Casein Kinase Ie and Id mRNAsin the Mouse Suprachiasmatic Nucleus

Yoshiki Ishida,1,2 Kazuhiro Yagita,1 Tsuyoshi Fukuyama,1 Masataka Nishimura,1

Mamoru Nagano,3 Yasufumi Shigeyoshi,3 Shun Yamaguchi,1 Takahide Komori,2 andHitoshi Okamura1*1Department of Brain Sciences, Kobe University Graduate School of Medicine, Kobe, Japan2Department of Oral and Maxillofacial Surgery, Kobe University School of Medicine, Kobe, Japan3Department of Anatomy, Kinki University School of Medicine, Osaka-Sayama, Japan

Casein kinase Ie (CKIe) and casein kinase Id (CKId) phos-phorylate clock oscillating mPER proteins, and play a keyrole in the transcription (post)translation feedback loopthat generates circadian rhythm. In the present study, theexpression profiles of CKIe and CKId mRNAs were ex-amined in the mice clock center, suprachiasmatic nu-cleus (SCN). Moderate levels of CKIe and CKId mRNAswere constantly expressed in the SCN in both light:darkand constant dark conditions. This finding supports thehypothesis that CKI may form a constant threshold to thenuclear entry of mPER proteins as in the Drosophilahomologue, double-time. Further, we demonstrated thatthe light exposure at subjective night induced a delayedincrease in CKIe and CKId mRNAs in the SCN. CKIe andCKId proteins may play a role on light-induced phase-shift. J. Neurosci. Res. 64:612–616, 2001.© 2001 Wiley-Liss, Inc.

Key words: circadian rhythm; casein kinase; clockgenes; suprachiasmatic nucleus

Molecular dissection of clock genes has revealed thatmammals and Drosophila share a common self-sustainedcore circadian clock oscillating system (Dunlap, 1999). Inboth Drosophila and mammals, the clock oscillating geneperiod (per) family is activated by the CLOCK/BMAL1(CYCLE) heterodimer (Darlington et al., 1998;Gekkakis et al., 1998). The nuclear translocation of neg-ative factors is regulated by the phosphorylation of PERby DOUBLETIME (DBT) in Drosophila (Kloss et al.,1998; Price et al., 1998). Constantly expressed DBT phos-phorylates cytoplasmic PER monomers and degradesthem (Kloss et al., 1998; Price et al., 1998), although theremaining unphosphorylated PER monomers dimerizewith TIM, and enter into the nucleus to suppress thetranscription of per. In mammals, Vielhaber et al. (2000)reported that casein kinase Ie (CKIe), a homologue of theDrosophila DBT, interacts with mPER1 and preventsmPER1 nuclear entry in cell lines in vitro. Recent geneticanalysis of the tau mutation, which is a semi-dominant

autosomal allele with a short period length in Syrianhamsters, has revealed that tau is a single gene mutation ofCKIe that modifies the ability of CKIe to phosphorylatePER in vitro (Lowrey et al., 2000).

In mammals, the circadian pacemaker is found in thesuprachiasmatic nucleus (SCN) located in the hypothala-mus. The expression profiles of casein kinases, however,have not been examined. In the present study, we inves-tigated the circadian expression of CKIe and the closelyrelated casein kinase d (CKId), which also have mPERphosphorylating potential (Vielhaber et al., 2000), in themouse SCN, and compared it to that in the liver. Becauseit is known that very high levels of mPer1 and mPer2 areacutely induced in light after brief light exposure, we alsotested whether or not CKIe and CKId are light-induciblegenes.

MATERIALS AND METHODS

Male Balb/c adult mice (Japan Animals Co., Osaka, Japan)were used for all experiments. Mice were housed for at least 2weeks of adaptation with the standard 12:12 hr light:dark (LD)cycle. In the DD cycle (constant dark) and light pulse experi-ments, we used mice on the second day of DD conditions. Thedaily expression of CKIe and CKId was examined every 4 hr(n 5 5 per time point) in the final LD cycle, starting at lights on(expressed as the Zeitgeber Time [ZT] 0), and in the secondcycle of constant darkness (DD) starting at subjective dawn(CircadianTime [CT] 0). For measuring the effect of lightstimulus (Takumi et al., 1998), we were exposed to an incan-

Contract grant sponsor: Special Coordination Funds of the Science andTechnology Agency of Japan; Contract grant sponsor: Grant-in-Aid for theScientific Research on Priority Areas of the Ministry of Education, Science,Sports and Culture of Japan; Contract grant sponsor: Mitsubishi Founda-tion; Contract grant sponsor: SRF.

*Correspondence to: Hitoshi Okamura, MD, PhD, Division of MolecularBrain Science, Department of Brain Sciences, Kobe University GraduateSchool of Medicine, 7-5-1 Kusunoki-cho, Chuo-ku, Kobe 650-0017,Japan. E-mail: okamurah@kobe-u.ac.jp

Received 16 January 2001; Revised 13 February 2001; Accepted 19 Feb-ruary 2001

Journal of Neuroscience Research 64:612–616 (2001)

© 2001 Wiley-Liss, Inc.

Fig. 1. CKIe and CKId mRNAs in the suprachiasmatic nucleus (SCN). A: In situ hybridization in the mouse brain using sense and antisenseprobes. Arrows indicate SCN. B: The expression profiles of CKIe (left panels) and CKId (right panels) mRNAs in the SCN in LD 12:12 (opencircles) and in DD (closed circles). For the CKIe and CKId genes, the relative RNA abundance was determined by quantitative in situhybridization, with the peak value adjusted to 100%. Values are expressed as means 6 SEM (n 5 5). Representative autoradiograms for each timepoint are showed above each graph. Numbers on each autoradiogram indicate the sampling time (hr). Scale bar 5 1 mm.

Casein Kinase in SCN 613

descent light stimulus (600 lux, 30 min) at CT4 and CT16 in thesecond DD cycle. Deeply anesthetized animals with ether weresacrificed 0, 30, 60, 120, and 240 min after the initiation of lightexposure, and processed for in situ hybridization of CKIe,CKId, and mPer1. All animal experiments were conducted inaccordance with the local Code of Practice (Kobe UniversitySchool of Medicine).

The 33P-UTP (New England Nuclear, Dupont, Boston,MA) labeled probes were the sense and antisense strands of CKIe(nucleotide positions 181–847), CKId (383–912) and mPer1(538–1,752) cRNAs. Coronal SCN sections (40 mm in thick-ness) made by a cryostat were processed for in situ hybridizationas described elsewhere (Shigeyoshi et al., 1997). The radioac-tivity of SCN on each section of a BioMax film (Kodak,Rochester, NY) was analyzed by an image analyzing system(MCID, Imaging Research Inc., Canada) after conversion intorelative optical densities produced by the 14C-autoradiographicmicroscales (Amersham, Buckinghamshire, UK), and total SCNcounts were summed. Statistical analyses of the data were madeusing one-way ANOVA followed by Scheffe’s multiple com-parisons.

For Northern blot analysis, mice were killed every 4 hrunder DD conditions. SCN from 10 mice was punched outfrom a brain slice (500 mm thick) with a microdissecting needle(i.d. 500 mm). The SCN and liver were quickly homogenized inTrizol Reagent (GIBCO BRL, Gaithersburg, MD), and theextracted total RNA of punched SCN (30 mg) and livers (20 mg)were electrophoresed in a 1.2% agarose gel containing 2%formaldehyde. RNAs were transferred to Biodyne Nylon Mem-branes (PALL Biosupport, New York) and hybridized withCKIe and CKId probes. G3PDH (Clontech, Palo Alto, CA) wasused as a control.

RESULTSTo test whether CKIe mRNA is actually expressed

in the master circadian pacemaking cells, we carried out insitu hybridization and Northern blot analysis in the SCN.We found a distinct CKIe mRNA signal in the SCN viain situ hybridization using an antisense probe, but nosignals with a sense probe (Fig. 1A). The signal level wasmoderate in the SCN, but the level was constant at alltimes both in LD (Fig. 1B) and DD (Fig. 1C). Consistentwith the in situ hybridization data, Northern blot analysis

also showed CKIe mRNA was expressed in the SCN withno rhythm in the DD condition (Fig. 2).

Then we examined the circadian profile of a closelyrelated CKI protein, CKId, which interacts with the SCN,using in situ hybridization and Northern blotting. CKIdmRNA was expressed at a slightly higher level in com-parison with CKIe mRNA in the SCN. This signal wasspecific because no labeling was detected when using senseprobes (Fig. 1D). Although the expression of CKIdmRNA was higher at ZT/CT12 than at ZT/CT0 inSCN, there was not statistically significant (ANOVA) (Fig.1E,F). Together with the Northern blot analysis showingconstant CKId and CKIe mRNA in the SCN (Fig. 2), weconclude that moderate levels of CKIe and CKId mRNAsare constantly expressed in the SCN in both LD and DDconditions.

Because it was recently found that rhythmic clockgenes were expressed in peripheral tissues, we examinedthe expression of CKIe mRNA and CKId mRNA in theliver by Northern blot analysis in DD (Fig. 2). In contrastto SCN, the expression of CKIe mRNA was very low,and did not show overt rhythm. The level of CKIdmRNA was moderate, but did not show rhythm in theliver.

Because some of the clock genes are light inducibleand are important for rhythmic phase-shift, we examinedCKIe and CKId mRNAs after a brief light exposure(30 min, 600 lux). When light was exposed at subjectivenight (CT16), when a phase-delay in locomotor activityoccurs in this strain of mice, CKIe transcripts began toincrease after 30 min, peaked at 120 min and returned tonear the basal level about 240 min after the pulse began(Fig. 3A). These delayed mRNA increases and the subse-quent decrease were slow compared with the rapidly in-ducible mPer1 gene, which showed an mRNA peak after60 min with rapid suppression (Fig. 3A). CKId transcriptlevels began to rise after 60 min after the start of theexposure to light, peaked at 120 min and declined grad-ually thereafter (Fig. 3B). The light influence was “gated”because the light-induced elevation of CKIe and CKIdmRNAs was no greater in the subjective daytime (CT6)(Fig. 3C,D).

Fig. 2. Northern blot analysis of CKIe (upper pan-els) and CKId (middle panels) in homogenates ofSCN (left panels) and the liver (right panels). Ex-periments were carried out at the indicated timepoints in the DD condition, and G3PDH was usedas a control.

614 Ishida et al.

DISCUSSIONIn the present study, we demonstrated that CKIe, a

possible phosphorylating protein of mammalian mPER1,was constantly expressed at moderate levels in both the LDand DD conditions in the SCN, where the master oscil-latory clock is localized. We also demonstrated that an-other CKI protein, CKId (97% identical to CKIe over thekinase domain), which is also capable of phosphorylatingmPER proteins (Vielhaber et al., 2000), was expressed inthe SCN, showing no rhythm in the LD and DD condi-tions. These observations suggest that both CKIe andCKId proteins may contribute to the phosphorylation ofmPER proteins in SCN cells. As for the tau mutation, itwas demonstrated that a single mutation of the CKIekinase domain produced a mutant protein having a po-

tential to bind mPER1 without phosphorylation activity(Lowrey et al., 2000). If both CKIe and CKId has thepotential to phosphorylate PER proteins by binding thespecific site, the reason why CKId cannot compensate formPER-phosphorylation in the tau mutation case might bethe dominant negative effect of mutated CKIe on theCKId to bind and phosphorylate mPERs. Which CKIplays a dominant role in phosphorylation step in thecircadian feedback loop must wait for future analyses ofthe genetic engineered mice of the both genes.

In Drosophila, constantly expressed DBT, a CKIehomologue, phosphorylates cytoplasmic PER monomers,and this tunes the nuclear translocation of PER/TIM.Kloss et al. (1998) reported that levels of dbt mRNA inadult heads showed no evident rhythm, and suggested that

Fig. 3. Time-course of the light exposure on CKIe (A,C), CKId (B,D) and mPer1 (A, dotted line) mRNAs in the SCN. The light pulse wasdelivered at 600 lux from CT16–CT16.5 (30 min; gray bar) (A,B) at CT4-CT4.5 (30 min; gray bar) (C,D). Relative RNA abundance wasdetermined with the value just before onset of light adjusted to 100% for CKIe, CKId, and mPer1. Values are expressed as mean 1 SEM (n 55). For CKIe , CKId; *P , 0.001, compared with the value just before light onset (ANOVA). Representative autoradiograms are shown on thegraph. Numbers on each autoradiogram indicate the sampling time (min) after the light onset.

Casein Kinase in SCN 615

the constant activity of the DBT sets a constant thresholdfor nuclear entry. The similarity of the expression profile,expression level, and substrate specificity suggests that thephosphorylation of mPER proteins by CKIe and CKIdcould be a critical step in forming the long period lengthof the negative feedback within the transcription-translation-based autoregulatory loop in mammals. Viel-haber et al. (2000) reported that CKIe interacts withmPER1 and prevents its nuclear entry into cell lines invitro. In mammals, it is now evident that both mRNA andproteins have robust rhythm, but the mPER proteinrhythm peaks delayed by about 6 hr compared with thoseof the mRNA cycles in both the LD and DD conditions(Hastings et al., 1999). This delayed appearance of mPER1and mPER2 proteins in the nucleus are thought to bederived from the functions of CKIe and CKId, whichinteract and degrade mPERs by phosphorylation.

It is well established that a single light pulse deliveredin the early night induces phase delay in locomotor activ-ity rhythms and high expression of mPer1 mRNA in theSCN (Shearman et al., 1997; Shigeyoshi et al., 1997).Although we showed the light-induced elevation of CKIeand CKId mRNAs in the SCN in subjective night, theinduction magnitudes were much smaller and the time-course was much more delayed (30–60 min) than thatseen in mPer1. If this delayed mRNA increase accompa-nies the subsequent increase of CKIe/CKId phosphoryla-tion activity, then it may be one of the cause of anunexpected small increase in mPER1 and mPER2 pro-teins in the nucleus compared with the strong induction ofthese mRNAs after the light exposure (Hastings et al.,1999; Field et al., 2000). Future studies of CKIe and CKIdproteins will address their roles of circadian oscillation andlight-induced phase-shift.

ACKNOWLEDGMENTSThis work was supported in part by grants from the

Special Coordination Funds of the Science and Technol-ogy Agency of Japan, Grant-in-Aid for the ScientificResearch on Priority Areas of the Ministry of Education,

Science, Sports and Culture of Japan, Ministry of Welfareof Japan, the Mitsubishi Foundation and SRF.

REFERENCESDarlington TK, Wager-Smith K, Ceriani MF, Staknis D, Gekakis N,

Steeves TDL, Weitz CJ, Takahashi JS, Kay SA. 1998. Closing thecircadian loop: CLOCK-induced transcription of its own inhibitors perand tim. Science 280:1599–1603.

Dunlap JC. 1999. Molecular bases for circadian clocks. Cell 96:271–290.Field MD, Maywood ES, O’Brien J, Weaver DR, Reppert S, Hastings

MH. 2000. Analysis of clock proteins in mouse SCN demonstratesphylogenic divergence of the circadian clockwork and resetting mecha-nisms. Neuron 25:437–447.

Gekakis N, Staknis D, Nguyen HB, Davis FC, Wilsbacher LD, King DP,Tahahashi JS, Weitz CJ. 1998. Role of the CLOCK protein in themammalian circadian mechanism. Science 280:1564–1569.

Hastings MH, Field MD, Maywood ES, Weaver DR, Reppert SM. 1999.Differential regulation of mPER1 and mTIM proteins in the mousesuprachiasmatic nuclei: new insights into a core clock mechanism. J Neu-rosci 11:1–7.

Kloss B, Price JL, Saez L, Blau J, Rothenfluh A, Wesley CS, Young MW.1998. The Drosophila clock gene double-time encodes a protein closelyrelated to human casein kinase Ie. Cell 94:97–107.

Lowrey PL, Shimomura K, Antoch MP, Yamazaki S, Zemenides PD,Ralph MR, Menaker M, Takahashi JS. 2000. Positional synthetic cloningand functional characterization of the mammalian circadian mutation tau.Science 288:483–492.

Price JL, Blau J, Rothenfluh A, Abodeely M, Kloss B, Young MW. 1998.Double-time is a new Drosophila clock gene that regulates PERIODprotein accumulation. Cell 94:83–95.

Shearman LP, Zylka MJ, Weaver DR, Kolakowski LFJ, Reppert SM. 1997.Two period homologs: circadian expression and photic regulation in thesuprachiasmatic nuclei. Neuron 19:1261–1269.

Shigeyoshi Y, Taguchi K, Yamamoto S, Takekida S, Yan L, Tei H, MoriyaT, Shibata S, Loros JJ, Dunlap JC, Okamura H. 1997. Light-inducedresetting of a mammalian circadian clock is associated with rapid induc-tion of the mPer1 transcript. Cell 91:1043–1053.

Takumi T, Matsubara C, Shigeyoshi Y, Taguchi K, Yagita K, MaebayashiY, Sakakida Y, Okumura K, Takashima N, Okamura H. 1998. A newmammalian period gene predominantly expressed in the suprachiasmaticnucleus. Genes Cells 3:167–176.

Vielhaber E, Eide E, Rivers A, Gao Z-H, Virshup DM. 2000. Nuclearentry of the circadian regulator mPER1 is controlled by mammaliancasein kinase 1e. Mol Cell Biol 20:4888–4899.

616 Ishida et al.