Constitutive expression of the Period1 gene impairsbehavioral and molecular circadian rhythmsRika Numano*†‡, Shin Yamazaki†§¶, Nanae Umeda†�, Tomonori Samura�, Mitsugu Sujino�, Ri-ichi Takahashi**,Masatsugu Ueda**, Akiko Mori††, Kazunori Yamada††, Yoshiyuki Sakaki*‡‡, Shin-Ichi T. Inouye�,Michael Menaker§, and Hajime Tei*§§

*Human Genome Center, Institute of Medical Science, University of Tokyo, 4-6-1 Shirokanedai, Minato-ku, Tokyo 108-8639, Japan; §Department of Biology,University of Virginia, Charlottesville, VA 22903-2477; �Department of Physics, Informatics, and Biology, Yamaguchi University, Yoshida, Yamaguchi753-8512, Japan; **Y. S. New Technology Institute, Inc., 1198-4 Utsunomiyashi, Iwaso-machi, Tochigi 321-0973, Japan; ††Mitsubishi Pharma Corporation,1000 Kamoshida-cho, Aoba-ku, Yokohama 227-0033, Japan; and ‡‡RIKEN Genomic Sciences Center, Human Genome Research Group, W402, 1-7-22Suehiro-cho, Tsurumi-ku, Yokohama 230-0045, Japan

Communicated by Joseph S. Takahashi, Northwestern University, Evanston, IL, January 4, 2006 (received for review March 25, 2005)

Three mammalian Period (Per) genes, termed Per1, Per2, and Per3,have been identified as structural homologues of the Drosophilacircadian clock gene, period (per). The three Per genes are rhyth-mically expressed in the suprachiasmatic nucleus (SCN), the centralcircadian pacemaker in mammals. The phases of peak mRNA levelsfor the three Per genes in the SCN are slightly different. Lightsequentially induces the transcripts of Per1 and Per2 but not of Per3in mice. These data and others suggest that each Per gene has adifferent but partially redundant function in mammals. To eluci-date the function of Per1 in the circadian system in vivo, wegenerated two transgenic rat lines in which the mouse Per1(mPer1) transcript was constitutively expressed under the controlof either the human elongation factor-1� (EF-1�) or the rat neuron-specific enolase (NSE) promoter. The transgenic rats exhibited an�0.6–1.0-h longer circadian period than their wild-type siblings inboth activity and body temperature rhythms. Entrainment inresponse to light cycles was dramatically impaired in the transgenicrats. Molecular analysis revealed that the amplitudes of oscillationin the rat Per1 (rPer1) and rat Per2 (rPer2) mRNAs were significantlyattenuated in the SCN and eyes of the transgenic rats. These resultsindicate that either the level of Per1, which is raised by overex-pression, or its rhythmic expression, which is damped or abolishedin over expressing animals, is critical for normal entrainment ofbehavior and molecular oscillation of other clock genes.

transgenic rats � entrainment

Physiological and behavioral circadian rhythms in mammalsare orchestrated by a central circadian clock located in the

suprachiasmatic nucleus (SCN) of the hypothalamus (1–4).Circadian rhythms persist under constant conditions and areentrained by environmental light cycles (5). Removal of the SCNcauses arrhythmicity of locomotor activity and transplant of fetalSCN tissue restores circadian rhythmicity with the period of thedonor SCN (4, 6, 7). Mammalian circadian photoreceptors arelocated in the retina, and specialized retinal ganglion cellscarrying light information project directly to the SCN (8, 9).

The mammalian Per1 gene has been identified as a structuralhomologue of the Drosophila circadian clock gene period (10,11). Two other homologues, Per2 and Per3, have also been foundin mammals (12–15). Both transcripts and proteins of all threePer genes exhibit circadian rhythms in the SCN. Per1 transcrip-tion is activated by the CLOCK–BMAL1 complex through itsbinding to E-box sequences located in the Per1 promoter region(16–18) and is repressed by Cryptochrome (Cry1 or Cry2) (19).Loss of function of the Clock (20, 21), Bmal1 (22), Cry1, or Cry2(23) genes affects not only the molecular oscillation of Per1 andPer2 but also circadian rhythmicities in behavior and the neuralactivity of the SCN (24–26). Per1 and Per2 expression is inducedimmediately after light exposure (27–30), and both light-inducedbehavioral phase shifts and glutamate-induced phase shifts in

SCN slice preparations are inhibited or attenuated by pretreat-ment with an antisense oligonucleotide against Per1 (31, 32).

To determine the role of the rhythmic expression of Per1 andits induction by light, we produced transgenic rat lines in whichthe transgene mPer1 was constitutively expressed under thecontrol of the human elongation factor-1� (EF-1�) or ratneuron-specific enolase (NSE) promoter. Constitutive expres-sion of the mPer1 transgene disrupted endogenous rPer1 andrPer2 oscillations, lengthened circadian period, and impairedlight entrainment.

ResultsGeneration and Characterization of Transgenic Rats OverexpressingmPer1. Transgenic rat lines in which the expression of mPer1 wasdriven by the promoter of either EF-1� or NSE developednormally and showed no obvious defects.

We examined the distribution of Per1 in the brains ofE12(Per1) and N3(Per1) rats by in situ hybridization. The Per1probe used in this experiment did not distinguish between thetranscripts of the endogenous rPer1 and the mPer1 transgene.Although a clear circadian fluctuation of rPer1 was observed inthe SCN from the wild-type siblings, no circadian oscillation ofthe Per1 signal (sum of the signals of both mPer1 and rPer1) wasdetected in the SCNs of E12(Per1) or N3(Per1) rats (Fig. 1B).Furthermore, in N3(Per1) rats strong signals, which we assumedto be N3(Per1) transcripts, were also detected in the whole brainas well as in the SCN (Fig. 1B). As expected, f luctuation of thePER protein was observed in the SCN of wild-type rats but notin the SCNs of E12(Per1) or N3(Per1) rats (Fig. 1C).

We then asked whether light exposure increased the amount ofPer1 mRNA in the SCN. Light exposure [30 min at circadian time(CT) 16] induced a 3-fold increase of Per1 transcript in the SCN inthe wild-type rats (Fig. 1D). In contrast, no Per1 induction wasobserved in the SCNs of E12(Per1) and N3(Per1) transgenic ratsafter light exposure (Fig. 1D). We also measured the light inductionof c-fos mRNA in the SCN by in situ hybridization and observed thata light pulse at CT 16 rapidly increased c-fos mRNA in bothwild-type and transgenic rats (Fig. 1E).

Conflict of interest statement: No conflicts declared.

Abbreviations: SCN, suprachiasmatic nucleus; NSE, neuron-specific enolase; EF-1�, elonga-tion factor-1�; CT, circadian time; LD, light�dark; DD, constant darkness.

†R.N., S.Y., and N.U. contributed equally to this work.

‡Present address: Department of Molecular and Cell Biology, University of California,Berkeley, CA 94720-3200.

¶Present address: Department of Biological Sciences, Vanderbilt University, VU Station B,Box 35-1634, Nashville, TN 37235-1634.

§§To whom correspondence should be sent at the present address: Research Group ofChronogenomics, Mitsubishi Kagaku Institute of Life Sciences, 11, Minami-Oya, Mahida,Tokyo 194-8511, Japan. E-mail: tei@libra.ls.m-kagaku.co.jp.

© 2006 by The National Academy of Sciences of the USA

3716–3721 � PNAS � March 7, 2006 � vol. 103 � no. 10 www.pnas.org�cgi�doi�10.1073�pnas.0600060103

Dow

nloa

ded

by g

uest

on

Aug

ust 1

9, 2

021

Behavioral and Physiological Phenotypes of mPer1 Transgenic Rats.We found altered circadian features in both the E12(Per1) andN3(Per1) lines. Representative body temperature rhythms fromboth transgenic lines and their wild-type siblings are shown inFig. 2 A and B. Both transgenic rat lines failed to entrain tolaboratory light�dark (LD) cycles (LD 12:12; �100–200 lux).Some masking was shown by both transgenic rat lines (the bodytemperature is low during the light period and becomes highimmediately after switching off the light), and ‘‘relative coordi-nation’’ (i.e., the period of the rhythm is modified as it crossesthe LD cycle) also occurred in both lines. As can be seen in Fig.2 A and B, progression of the high-temperature phase relative tothe LD cycle was slower when a high temperature coincided withdarkness and faster when it coincided with light. As a conse-quence, the high temperature tended to remain in the darkportion of the LD cycle before transiting quickly through the

light portion. This observation suggests that the environmentallight signal still reached the pacemaker of the transgenic rats,although it was too weak to fully entrain the rhythms. Relativecoordination in locomotor activity rhythm was not as clear as inbody temperature rhythms because of the stronger maskingeffect of light on the wheel running (see Fig. 4, which is publishedas supporting information on the PNAS web site). The free-running periods of body temperature rhythms in constant dark-ness (DD) from the male F1 E12(Per1) and N3(Per1) transgenicrats were 0.82 h and 0.95 h longer than from their wild-typesiblings, respectively. Periods in individual rats were: 25.3, 25.3,and 25.2 h for E12(Per1); 24.4 and 24.5 h for wild-type siblingsof E12(Per1); 25.3 h for N3(Per1); and 24.3 and 24.4 h forwild-type siblings of N3(Per1). Two other F1 transgenic rat linessimilarly showed longer periods in wheel-running rhythms thantheir wild-type siblings (data not shown).

The period distributions of wheel-running rhythms in theoffspring of F1 heterozygous and wild-type rats are shown in Fig.2C. Approximately 50% of pups from these crosses carried thetransgene. The distribution of the free-running periods of thetransgenic rats and of their wild-type siblings did not overlap(Fig. 2C). The average circadian periods of the transgenic ratswere 24.92 � 0.03 (SEM) [E12(Per1)] and 24.99 � 0.02[N3(Per1)] and were significantly longer than those of theirwild-type siblings [24.29 � 0.01 for wild-type siblings ofE12(Per1); 24.35 � 0.03 for wild-type siblings of N3(Per1)](Fig. 2C).

Expression of mPer1, rPer1, and rPer2 in SCN and the Eyes of mPer1Transgenic Rats. The rPer1 mRNA in the SCN of the wild-type ratswas rhythmic, with a peak between CT 4–8 (Fig. 3 and Table 1).In the SCN of E12(Per1) rats, the mPer1 transgene was consti-tutively expressed at approximately twice the peak level of therPer1 transcript in the wild-type rats. In the SCN of N3(Per1)rats, the mPer1 transgene was constitutively expressed at �7times the peak level of the rPer1 transcript in the SCN of thewild-type rats (Fig. 3 A and E). The robustness of endogenousrPer1 mRNA cycling in the SCN of E12(Per1) and N3(Per1) ratswas significantly attenuated compared with those in the wild-type rats (Fig. 3 A and E and Table 1).

We also measured the levels of the rPer2 transcripts in thesame samples and found that rPer2 had expression profilessimilar to those of rPer1. In the SCN of the wild-type rats, rPer2mRNA showed a robust circadian pattern with a peak at CT 12,but the robustness of the oscillation was attenuated in the SCNof E12(Per1) and N3(Per1) rats [with the exception of theresidual rhythmicity in the SCN of E12(Per1) rats] (Fig. 3 C andG and Table 1).

Similar effects were observed in the mRNA extracted fromeyes (Fig. 3 B, D, F, and H). The mPer1 transgene was consti-tutively expressed in the eyes of E12(Per1) and N3(Per1) rats at�3 and 4 times the peak level of the rPer1 transcript in wild-typerats. The amplitudes of oscillation of rPer1 and rPer2 mRNAlevels in the eyes of both E12(Per1) and N3(Per1) rats were muchlower than that of wild-type rats, and residual rhythmicities ofthe rPer1 and rPer2 transcripts were observed only in E12(Per1)rats (Fig. 3 B, D, F, and H and Table 1).

DiscussionTwo transgenic rat lines in which the mPer1 transgene wasconstitutively expressed under the control of either the EF-1� orNSE promoter showed significantly lengthened circadian peri-ods and abnormal entrainment of the rhythms of locomotoractivity and body temperature. The amplitude of the circadianrhythms of endogenous rPer1 and rPer2 was severely attenuatedin these transgenic rats. That the phenotypes of both transgenicrat lines were almost identical can be explained if Per1 overex-pression is assumed to be saturating despite the different copy

Fig. 1. Structures of EF-1�::mPer1 and NSE::mPer1 transgenes and histolog-ical analysis of SCN of transgenic rat lines. (A) mPer1 cDNA was sited betweenthe human EF-1� or rat NSE::Per1 promoters and SV40 poly(A) signals. Primers(arrows) and probes (white boxes) used for the screening of the transgenic ratlines are indicated below each construct. (B) Per1 mRNA levels in coronalsections of the brain of the wild-type, E12(Per1), and N3(Per1) rat lines wereanalyzed by in situ hybridization. Per1 transcripts were constitutively ex-pressed in the SCN of the E12(Per1) and the N3(Per1) transgenic rats lines,whereas clear circadian cycling of rPer1 expression was observed in the SCN ofthe wild-type rat lines. (C) Immunohistochemistry of PER1 protein in SCN oftransgenic rat lines at CT 0 and CT 12. Photomicrographs of SCN and otherbrain regions after staining with the PER1 antibody. (D) Light induction of Per1mRNA in the SCN of wild-type, E12(Per1), and N3(Per1) rats was compared byin situ hybridization analysis. Per1 mRNA was not induced in the SCN of theE12(Per1) and N3(Per1) transgenic rat lines after a 30-min light exposure at CT16. Results are quantified in the histogram. (E) Light induction of c-fos mRNAin the SCN of wild-type, E12(Per1), and N3(Per1) rat lines. (Lower and Upper)Illustrated is the SCN with and without a light pulse at CT 16, respectively.

Numano et al. PNAS � March 7, 2006 � vol. 103 � no. 10 � 3717

GEN

ETIC

S

Dow

nloa

ded

by g

uest

on

Aug

ust 1

9, 2

021

numbers of the mPer1 transgene and the different levels of mPer1expression in the two lines. From these results, we conclude thatthe similar phenotypes in the two transgenic rat lines are due tothe constitutive expression of mPer1 rather than due to disrup-tion of the genes surrounding the insertion sites of the trans-

genes, and furthermore, we conclude that either the level of Per1,which is raised by overexpression, or its rhythmic expression,which is damped or abolished in over expressing animals, iscritical for normal entrainment of behavior and molecularoscillation of other clock genes. The data indicate that the light

Fig. 2. Body temperature and wheel-running activity rhythms of transgenic rats. Body temperature rhythms of E12(Per1) (A) and N3(Per1) (B) rats were reducedto actogram form (33) (shown double plotted) and compared with those of wild-type siblings. White and black bars at the top of figures indicate the LD cyclewhen present. Arrows indicate transitions from LD to DD. Both transgenic rat lines failed to entrain completely and, in DD, exhibited 0.82 h [E12(Per1)] and 0.95 h[N3(Per1)] longer circadian periods of body temperature than their wild-type siblings. General activity rhythms monitored by movement of the transmitter inthese rats are shown in Fig. 4. (C) The period distribution of wheel-running rhythms in the offspring of crosses between F1 heterozygous transgenic rats of bothlines and wild-type rats are plotted. ■ , wild-type rats; �, transgenic rats.

Fig. 3. Expression of Per1 and Per2 in SCN and eyes (whole eyes) of transgenic rat lines. The amounts of the rPer1, rPer2, and mPer1 mRNAs were quantifiedwith TaqMan RT-PCR analysis and are plotted as relative mRNA abundance. Each point represents mean � SEM values of four independent assays (n � 4). (A)Expression of mPer1 and rPer1 in the SCN of E12(Per1) and wild-type rats. (B) Expression of mPer1 and rPer1 in the eyes of E12(Per1) and wild-type rats. Expressionof rPer2 mRNA in the SCN (C) and eyes (D) of E12(Per1) and wild-type rats. (E) Expression of mPer1 and rPer1 in the SCN of N3(Per1) and the wild-type rats. (F)Expression of mPer1 and rPer1 in the eyes of N3(Per1) and wild-type rats. Expression of rPer2 mRNA in the SCN (G) and eyes (H) of N3(Per1) and wild-type rats.‚, mPer1; �, rPer1 in transgenic rats; �, rPer1 in wild-type rats (A, B, E, and F); �, rPer2 in transgenic rats; �, rPer2 in wild-type rats (C, D, G, and H).

3718 � www.pnas.org�cgi�doi�10.1073�pnas.0600060103 Numano et al.

Dow

nloa

ded

by g

uest

on

Aug

ust 1

9, 2

021

induction and rhythmic oscillation of Per1 expression in the SCNare strongly correlated with the entrainment and molecularoscillation.

In Drosophila, constitutive expression of per does not eliminatecyclic PER expression and behavioral rhythms, although themechanism that maintains PER cycling in the absence of rhyth-mic per transcription is still not completely understood (34). Asimilar posttranscriptional regulation of PER expression mightbe expected in mammals because the sequences in the 3�UTR ofmPer1 inhibit the translation of mPer1 mRNA (35). Despite thedifferences between rats and mice, cross-species comparisons ofthe phenotypes resulting from overexpression or absence ofexpression of Per genes are of significant interest (36–40).Per1�/� and Per2�/� mice exhibit shortened periods of wheel-running activity. The double knockout (Per1�/� Per2�/�) isarrhythmic (36, 37). These results indicate that loss of expressionof either Per1 or Per2 does not abolish the circadian oscillationbut does influence the length of its period. Because the Per1�/�

Per2�/� mice are completely arrhythmic, it has been suggestedthat there is partial redundancy in the function of the Per1 andPer2 genes (36, 37). Therefore, it seems that posttranscriptionalregulation of Per genes and functional redundancy of the Per1and Per2 genes may explain the residual behavioral rhythmobserved in our transgenic animals. In addition, because thePer1�/� mouse has a shortened period (36, 38) and because weobserved lengthening in the period in our Per1 transgenic rats,the period of the rhythm may be positively correlated with thegene dosage of Per1. Thus, future construction of transgenicanimals in which the expression of Per1 is clamped at low,middle, and high levels and those in which both Per1 and Per2 areoverexpressed may clarify the roles of these genes in the deter-mination and generation of circadian rhythms.

More dramatic than the period phenotype of our transgenicrats is the defect in entrainment. There are three possibleexplanations for this effect of mPer1 overexpression: (i) Thetransgenic rats may be less sensitive to the entraining lightstimulus. This hypothesis seems unlikely because our data dem-onstrate rapid c-fos induction in the SCN of the transgenic ratsafter a light pulse (Fig. 1E). Thus, the light induction of rPer1must be obscured by the mPer1 overexpression in the transgenicrats (Fig. 1D). (ii) The lengthened period may place the circadianpacemaker outside the limits of entrainment without affectinglight entrainment mechanisms. To test this possibility, we sub-

jected wild-type rats (n � 5) to 23-h light cycles (LD 11.5:11.5;100 lux) and found that they could entrain (S.Y. and N.Schneider, unpublished data). Because entrainment to this cyclerequires a larger phase shift each day (�1.5-h advance) than the�1-h advance that would be required to entrain the transgenicrats to the 24-h cycle (to which they failed to entrain), weconclude that the lengthened period of the transgenic ratscannot by itself account for the entrainment defect. (iii) A directeffect of excess mPer1 on the oscillator or a combination of sucha direct effect with reduced light sensitivity and�or lengthenedperiod could cause the circadian oscillator of Per1-overexpress-ing rats to be unable to advance �1 h�day as would be requiredfor entrainment. Although Albrecht et al. (39) reported that Per1mutant mice fail to phase advance in response to a 15-min lightexposure at Zeitgeber time 22 on the first night in DD and Per2mutant mice fail to phase delay at Zeitgeber time 14, Cermakianet al. (38) found that Per1 mutant mice show a normal phaseresponse to a 30-min light exposure at CT 14 or 20 after 3–5 daysin DD. Bae and Weaver (40) in a T-cycle photoperiod experi-ment concluded that both Per1 and Per2 mutant mice have theability to shift in both directions. Moreover, antisense studiessuggest that Per1 is involved in light-induced phase delays (31, 32,41). Because Per1 and Per2 interact with each other, overexpres-sion of mPer1 in transgenic rats could affect the mechanisminvolved in both phase advances and delays.

The present study shows that oscillations of both endogenousrPer1 and rPer2 expression in the SCN are impaired in mPer1-overexpressing transgenic rats. Per1 and Per2 transcription isinduced by the CLOCK–BMAL1 complex (16–18) and re-pressed by CRY1 or CRY2 (19). Although each of the PERproteins interacts with the CRY1 and CRY2 proteins, overex-pression of PERs in mouse NIH 3T3 cells has almost no criticaleffect on the transcription of Per1 and Per2 (18, 19). Thisdiscrepancy may be due to the difference in methods used in ourstudy (transgenic) and others (transient expression). An exten-sive elucidation of the molecular function of PER is necessary forfurther understanding of the regulation of Per1 transcription.

Materials and MethodsAnimals. All of the transgenic rat lines used in this study weregenerated and maintained by using the Wistar rat strain (CharlesRiver Laboratories Japan). To overexpress mPer1, we made twodifferent DNA constructs. The promoter regions of the human

Table 1. Cosinor summary of rhythm variations in transgenic rats

Transgenic line Tissue Gene Animals Amplitude P Acrophase degrees

E12(Per1) SCN rPer1 Wild type 0.737 0.0028 �111E12(Per1) 0.126 0.2372 �96

rPer2 Wild type 0.555 0.0123 �162E12(Per1) 0.191 0.0001 �136

Eye rPer1 Wild type 0.428 0.0082 �137E12(Per1) 0.359 0.0001 �133

rPer2 Wild type 0.29 0.0001 �170E12(Per1) 0.15 0.0011 �154

N3(Per1) SCN rPer1 Wild type 0.566 0.0018 �101N3(Per1) 0.144 0.5361 �169

rPer2 Wild type 0.512 0.0074 �160N3(Per1) 0.168 0.2134 �106

Eye rPer1 Wild type 0.409 0.0236 �135N3(Per1) 0.0948 0.5813 �165

rPer2 Wild type 0.498 0.0003 �151N3(Per1) 0.323 0.0876 �117

Chronobiometric values and parameters were statistically measured by the cosinor method by using TIME SERIES

SINGLE COSINOR 6.2 software (Laboratory of Applied Statistics and BioMedical Computing Expert Soft Technologie,Richelieu, France).

Numano et al. PNAS � March 7, 2006 � vol. 103 � no. 10 � 3719

GEN

ETIC

S

Dow

nloa

ded

by g

uest

on

Aug

ust 1

9, 2

021

EF-1� and rat NSE genes were directly ligated to the mPer1 ORFflanked by transcriptional termination signals derived from theSV40 early gene and were designated as EF-1�::mPer1 andNSE::mPer1, respectively (Fig. 1 A). We confirmed the expres-sion of the mPer1 mRNA and mPer1 protein in Chinese hamsterovary cells (JCRB9018, Japanese Collection of Research Biore-sources Cell Bank, Osaka) and PC12 cells (rat adrenal pheo-chromocytoma cells; IFO50015, Health Science Research Re-sources Bank, Osaka) 24 h after the transfection with these twoplasmids (data not shown). The two transgenic vectors werelinearized by using either SmaI�NotI (for EF-1�::mPer1) orHindIII�SacII (for NSE::mPer1) and injected into fertilizedWistar rat zygotes. Transgenic rats were screened by PCR andSouthern blot analyses by using their tail DNA. We obtained 17and 10 founder rats in the EF-1�::mPer1 and NSE::mPer1transgenic experiments, respectively. Two founder rats ofEF-1�::mPer1 transgenic lines were mosaic and four founder ratsof NSE::mPer1 transgenic lines were either mosaic (three lines)or sterile (one line). In the initial screening, we measured thewheel-running activity of the F1 heterozygous transgenic ratsand compared it with that of their wild-type siblings. We foundthat two lines of the EF-1�::mPer1 transgenic rats and one lineof the NSE::mPer1 transgenic rats showed free-running periodssignificantly longer than those of wild-type rats in DD. In thisstudy, we used a line designated E12(Per1) from theEF-1�::mPer1 transgenic rats and a line designated N3(Per1)from the NSE::mPer1 transgenic rats to analyze molecular andphysiological circadian rhythms. The copy numbers of the mPer1transgenes in the rat genomes in these two lines were estimatedby Southern blot and TaqMan RT-PCR (Applied Biosystems)analyses. We found that approximately five copies of the trans-gene were inserted in the genome of E12(Per1) rats and 40copies in the genome of N3(Per1) rats. All of the animal studiesconducted were in accordance with the guidelines of the Com-mittee on Animal Care and Use of University Tokyo, Universityof Virginia, Yamaguchi University, and Y. S. New TechnologyInstitute, Inc.

Behavioral Analysis. F1 heterozygous transgenic rats and theirwild-type siblings (8 weeks old) were individually housed in cageswith running wheels (8 cm wide with a 17-cm diameter). Wheelrevolutions were monitored with wheel-activated micro switchesevery 6 min for 8 days in LD (12 h of light and 12 h of dark) and19 days in DD by using the Data Quest system (Data SciencesInternational, Arden Hills, MN). Using this behavioral screen,we found that F1 heterozygous E12(Per1) and N3(Per1) ratsshowed free-running periods �45 min longer than those of theirwild-type siblings. Therefore, we investigated more detailedcircadian characteristics in these rats [three E12(Per1) heterozy-gous males and two wild-type male siblings; one N3(Per1)heterozygous male and two wild-type male siblings]. Bodytemperature was measured by telemetry by using implantedtransmitters (model VM-FH; Mini-Mitter, Sunriver, OR). Aradio receiver (model RA-1010; Mini-Mitter) coupled to a dataacquisition system (Dataquest III; Data Sciences International)was placed under the cage. Body temperature was recorded in6-min bins and analyzed by ACTIVIEW (Mini-Mitter).

F1 female heterozygous transgenic rats [five females inE12(Per1) and two females in N3(Per1)] were crossed withwild-type rats. We obtained 25 E12(Per1) heterozygous (14males and 11 females) and 30 wild-type (17 males and 13females) siblings and 15 N3(Per1) heterozygous (9 males and 6females) and 15 wild-type (9 males and 6 females) siblings. Whenthe rats reached 4 weeks of age, they were individually housedin running cages, and their wheel-running activities were mon-itored for 14 days in LD and 10 days in DD. Free-running periodswere obtained on the basis of the eye fitted line, which markedthe onset of wheel-running activity during 10 days in DD. Tissues

for mRNA analysis were sampled from these rats after 10 daysin DD.

Cages were changed every 2 weeks and water bottles everyweek. Half of the bedding was replaced between each cagechange. All animal care in DD was performed under infraredillumination by using an infrared viewer (FJW Optical Systems,Palatine, IL). Ambient temperature in the cage was �21°C.

Quantification of mPer1, rPer1, rPer2, and GAPDH Transcripts. Theanimals produced by crossing the F1 heterozygous and wild-typerats were raised under the LD conditions as described above. Forthe N3(Per1) group, we did additional sampling (12 transgenicand 12 wild type), and these samples are included in the dataanalysis in Fig. 3. After 10 days in DD, activity onset (CT 12) waspredicted on the basis of the eye fitted line that marked the onsetof wheel-running activity during the 10-day period. Brainsand eyes (entire eye) were sampled at 4 h intervals starting atCT 0 of the 11th cycle in DD. By using an infrared viewer, therats were killed with CO2. Both eyes were removed with scis-sors without visible light exposure and frozen in liquid nitrogen.The brain was removed under light and frozen immediatelyon dry ice. Paired SCNs were punched out from a 1-mm-thickcoronal slice by using a microdissecting needle (inner diameter,600 �m). Total RNA from the SCN and eyes was extracted byusing TRIzol solution (Invitrogen), treated with DNase I (Strat-agene), and further purified with TRIzol LS solution (Invitro-gen). The mPer1, rPer1, rPer2, and GAPDH transcripts werequantified by using the TaqMan RT-PCR kit (Applied Biosys-tems) and a 7700 Sequence Detection System (Applied Biosys-tems). Serial dilutions of the rat genomic DNA solutions (from0.64 to 10,000 copies per �l) (Promega) were used as standardsof the copy numbers for the rPer1, rPer2, and GAPDH cDNAs.To quantify the mPer1 cDNA, the EF1�::mPer1 DNA solutionwas mixed with the rat genomic DNA solution to obtain standardsolutions containing equal numbers of copies of each DNA(from 0.64 to 10,000 copies per �l). The transcripts of theendogenous rPer1 gene and mPer1 transgene were quantified byusing specially designed primer sets; the forward and reverseprimers for rPer1 were designed within a unique 3�-untranslatedregion of rPer1; the forward primer for mPer1 was designed in the3� terminal of the mPer1 ORF and the reverse primer in poly(A)signals of SV40 (Fig. 1 A). The specific primer pairs used in thisstudy were: rPer1 forward, 5�-AGCAGAGTGGAAGTTT-TCAGCC-3�; rPer1 reverse, 5�-ACCACTTCAGCAGCTTGT-CAGC-3�; mPer1 forward, 5�-ACTCTGCCATGGAGGAA-GAAGA-3�; mPer1 reverse, 5�-TGTGAAATTTGTGATGC-TATTGCTT-3�; rPer2 forward, 5�-GGTGTGGCAGCTTTT-GCTTC-3�; rPer2 reverse, 5�-CGGCACAGAAACGTACAGT-GTG-3�.

The GAPDH transcript was used for normalizing the expres-sion of each transcript. The amplitudes of the rhythms werecalculated by cosinor analysis (TIME SERIES SINGLE COSINOR 6.2;Laboratory of Applied Statistics and BioMedical ComputingExpert Soft Technologie, Richelieu, France), and the results arepresented in Table 1.

In Situ Hybridization. The preparation and sampling of the animals(n � 3) were described above. After 14 days in DD, activity onsetwas predicted on the basis of the eye fitted line that marked theonset of wheel-running activity during the 14-day period. Brainswere sampled 60 min after a 30-min light exposure at CT 16.Light was produced by a 40 W fluorescent lamp [60 �W�cm2

(�200 lux) at cage level]. In situ hybridization analysis of Per1and c-fos transcripts was performed as described in ref. 28.Brains were fixed in 4% paraformaldehyde�PBS solution for24 h, followed by immersion in 20% sucrose�PBS solution for48 h. Brains were then sliced into 30-�m-thick sections with aCryostat (Leitz CM-1510). The sections were hybridized with the

3720 � www.pnas.org�cgi�doi�10.1073�pnas.0600060103 Numano et al.

Dow

nloa

ded

by g

uest

on

Aug

ust 1

9, 2

021

33P-labeled antisense cRNA probe complementary to the se-quence from nucleotides 436 to 931 of mPer1 or complementaryto the sequence from nucleotides 513 to 1021 of c-fos. Briefly,tissue sections were incubated with proteinase K (2 �g�ml in 10mM Tris�HCl, pH 7.5�0.5 M EDTA) for 10 min at 37°C andacetic anhydride in 0.1 M triethanolamine for 10 min. Then theywere incubated in hybridization buffer (60% formamide�10%dextran sulfate�1 M Tris�HCl, pH 7.4�5 M NaCl�0.5 M EDTA�500 �g/ml tRNA�10% SDS�1 � Denhardt’s solution) contain-ing mPer1 or c-fos antisense cRNA probes for 17–18 h at 60°C.After the posthybridization wash and RNaseA treatment, hy-bridized sections and 14C-acrylic standards (Amersham Phar-macia Biosciences) were exposed to BioMax film (Kodak)for 7 days.

Immunohistochemistry. The preparation and sampling of theanimals were described above. Rats were transferred to DDconditions 2 days before the experiment and killed at CT 0 andCT 12. Anti-mouse PER1 antibody was purchased from AffinityBioReagents (catalog no. PA1-524). After postfixation in para-formaldehyde and dehydration, brains were sectioned into 20-

�m-thick slices with a Cryostat microtome (Leica CM1510). Thesections were incubated with a polyclonal rabbit antibody againstPER1 diluted 1:1,000 in PBS containing 0.3% Triton X-100 for72 h at 4°C after the inactivation of endogenous peroxidase with0.3% H2O2�50% methanol. Then they were incubated withanti-rabbit IgG diluted 1:1,000 in PBS for 24 h at 4°C and withavidin–biotin peroxidase complex (Vectastain Elite ABC kit;Vector Laboratories), and peroxidase was visualized by using0.01% diaminobenzidine as chromogen and 0.01% H2O2 in PBSas substrate for 10 min.

We thank Drs. K. Hanaoka, K. Sakimura, and Y. Kimura for the gift ofthe plasmid containing EF-1� and NSE promoters and poly(A) signalsof the SV40 early gene, respectively; Dr. J. Eguchi for excellent technicalassistance in the TaqMan RT-PCR analysis; Drs. K. Nakao andM. Katsuki for helpful discussions; and Dr. C. Green for the criticalreading of the manuscript. This work was partially supported by researchgrants from the Japan Society for the Promotion of Science, the JapaneseMinistry of Education, Science, Sports and Culture, and the JapaneseMinistry of Health and Welfare, and National Institute of Mental HealthGrant MH56647 (to M.M.).

1. Pittendrigh, C. S. (1993) Annu. Rev. Physiol. 55, 16–54.2. Davidson, A. J., Yamazaki, S. & Menaker, M. (2003) Novartis Found. Symp.

253, 110–121.3. Yoo, S. H., Yamazaki, S., Lowrey, P. L., Shimomura, K., Ko, C. H., Buhr, E. D.,

Siepka, S. M., Hong, H. K., Oh, W. J., Yoo, O. J., et al. (2004) Proc. Natl. Acad.Sci. USA 101, 5339–5346.

4. Ralph, M. R., Foster, R. G., Davis, F. C. & Menaker, M. (1990) Science 247,975–978.

5. Takahashi, J. S. (1995) Annu. Rev. Neurosci. 18, 531–553.6. Sujino, M., Masumoto, K., Yamaguchi, S., van der Horst, G. T., Okamura, H.

& Inouye, S. T. (2003) Curr. Biol. 13, 664–668.7. Silver, R., LeSauter, J., Tresco, P. A. & Lehman, M. N. (1996) Nature 382,

810–813.8. Menaker, M. (2003) Science 299, 213–214.9. Foster, R. G. (2005) Nature 433, 698–699.

10. Tei, H., Okamura, H., Shigeyoshi, Y., Fukuhara, C., Ozawa, R., Hirose, M. &Sakaki, Y. (1997) Nature 389, 512–516.

11. Sun, Z. S., Albrecht, U., Zhuchenko, O., Bailey, J., Eichele, G. & Lee, C. C.(1997) Cell 90, 1003–1011.

12. Albrecht, U., Sun, Z. S., Eichele, G. & Lee, C. C. (1997) Cell 91, 1055–1064.13. Shearman, L. P., Zylka, M. J., Weaver, D. R., Kolakowski, L. F. & Reppert,

S. M. (1997) Neuron 19, 1261–1269.14. Takumi, T., Taguchi, K., Miyake, S., Sakakida, Y., Takashima, N., Matsubara,

C., Maebayashi, Y., Okumura, K., Takekida, S., Yamamoto, S., et al. (1998)EMBO J. 17, 4753–4759.

15. Zylka, M. J., Shearman, L. P., Weaver, D. R. & Reppert, S. M. (1998) Neuron20, 1103–1110.

16. Gekakis, N., Stankis, D., Nguyen, H. B., Davis, F. C., Wilsbacher, L. D., King,D. P., Takahashi, J. S. & Weitz, C. J. (1998) Science 280, 1564–1569.

17. Hogenesch, J. B., Gu, Y. Z., Jain, S. & Bradfield, C. A. (1998) Proc. Natl. Acad.Sci. USA 95, 5474–5479.

18. Hida, A., Koike, N., Hirose, M., Hattori, M., Sakaki, Y. & Tei, H. (2000)Genomics 65, 224–233.

19. Kume, K., Zylka, M. J., Sriram, S., Shearman, L. P., Weaver, D. R., Jin, X.,Maywood, E. S., Hastings, M. H. & Reppert, S. M. (1999) Cell 98, 193–205.

20. King, D. P., Zhao, Y., Sangoram, A. M., Wilsbacher, L. D., Tanaka, M.,Antoch, M. P., Steeves, T. D., Vitaterna, M. H., Kornhauser, J. M., Lowrey,P. L., et al. (1997) Cell 89, 641–653.

21. Antoch, M. P., Song, E. J., Chang, A. M., Vitaterna, M. H., Zhao, Y.,Wilsbacher, L. D., Sangoram, A. M., King, D. P., Pinto, L. H. & Takahashi, J. S.(1997) Cell 89, 655–667.

22. Bunger, M. K., Wilsbacher, L. D., Moran, S. M., Clendenin, C., Radcliffe, L. A.,Hogenesch, J. B., Simon, M. C., Takahashi, J. S. & Bradfield, C. A. (2000) Cell103, 1009–1017.

23. van der Horst, G. T., Muijtjens, M., Kobayashi, K., Takano, R., Kanno, S.,Takao, M., de Wit, J., Verkerk, A., Eker, A. P., van Leenen, D., et al. (1999)Nature 398, 627–630.

24. Herzog, E. D., Takahashi, J. S. & Block, G. D. (1998) Nat. Neurosci. 1, 708–713.25. Nakamura, W., Honma, S., Shirakawa, T. & Honma, K. (2002) Nat. Neurosci.

5, 399–400.26. Albus, H., Bonnefont, X., Chaves, I., Yasui, A., Doczy, J., van der Horst, G. T.

& Meijer, J. H. (2002) Curr. Biol. 12, 1130–1133.27. Zheng, B., Larkin, D. W., Albrecht, U., Sun, Z. S., Sage, M., Eichele, G., Lee,

C. C. & Bradley, A. (1999) Nature 400, 169–173.28. Shigeyoshi, Y., Taguchi, K., Yamamoto, S., Takekida, S., Yan, L., Tei, H.,

Moriya, T., Shibata, S., Loros, J. J., Dunlap, J. C. & Okamura, H. (1997) Cell91, 1043–1053.

29. Yan, L., Takekida, S., Shigeyoshi, Y. & Okamura, H. (1999) Neuroscience 94,141–150.

30. Wilsbacher, L. D., Yamazaki, S., Herzog, E. D., Song, E. J., Radcliffe, L. A.,Abe, M., Block, G., Spitznagel, E., Menaker, M. & Takahashi, J. S. (2002) Proc.Natl. Acad. Sci. USA 99, 489–494.

31. Akiyama, M., Kouzu, Y., Takahashi, S., Wakamatsu, H., Moriya, T., Maetani,M., Watanabe, S., Tei, H., Sakaki, Y. & Shibata, S. (1999) J. Neurosci. 19,1115–1121.

32. Tischkau, S. A., Mitchell, J. W., Tyan, S. H., Buchanan, G. F. & Gillette, M. U.(2003) J. Biol. Chem. 278, 718–723.

33. Refinetti, R. & Menaker, M. (1992) Physiol. Behav. 51, 613–637.34. Zeng, H., Hardin, P. E. & Rosbash, M. (1994) EMBO J. 13, 3590–3598.35. Kojima, S., Hirose, M., Tokunaga, K., Sakaki, Y. & Tei, H. (2003) Biochem.

Biophys. Res. Commun. 301, 1–7.36. Bae, K., Xiaowei, J., Maywood, E. S., Hastings, M. H., Reppert, S. M. & Weaver

D. R. (2001) Neuron 30, 525–536.37. Zheng, B., Albrecht, U., Kaasik, K., Sage, M., Lu, W., Vaishnav, S., Li, Q., Sun,

Z. S., Eichele, G., Bradley, A. & Lee, C. C. (2001) Cell 105, 683–694.38. Cermakian, N., Monaco, L., Pando, M. P., Dierich, A. & Sassone Corsi, P.

(2001) EMBO J. 20, 3967–3974.39. Albrecht, U., Zheng, B., Larkin, D., Sun, Z. S. & Lee, C. C. (2001) J. Biol.

Rhythms 16, 100–104.40. Bae, K. & Weaver, D. R. (2003) J. Biol. Rhythms 18, 123–133.41. Wakamatsu, H., Takahashi, S., Moriya, T., Inouye, S. T., Okamura, H.,

Akiyama, M. & Shibata, S. (2001) NeuroReport 12, 127–131.

Numano et al. PNAS � March 7, 2006 � vol. 103 � no. 10 � 3721

GEN

ETIC

S

Dow

nloa

ded

by g

uest

on

Aug

ust 1

9, 2

021