Voltar
Artigo Acesso aberto Revisado por pares
BIBTEX RIS

Two-peaked Synchronization in Day/Night Expression Rhythms of the Fibrinogen Gene Cluster in the Mouse Liver

2003; Elsevier BV; Volume: 278; Issue: 33 Linguagem: Inglês

10.1074/jbc.m304809200

ISSN

1083-351X

Autores

Eiko Sakao, Akinori Ishihara, Kazumasa Horikawa, Masashi Akiyama, Makoto Arai, Masaki Kato, Naohiko Seki, Kohji Fukunaga, Atsuko Shimizu-Yabe, Katsuro Iwase, Satoko Ohtsuka, Takeyuki Sato, Yoichi Kohno, Shigenobu Shibata, Masaki Takiguchi,

Tópico(s)

Adipose Tissue and Metabolism

Resumo

Genes expressed with day/night rhythms in the mouse liver were searched for by microarray analysis using an in-house array harboring mouse liver cDNAs. The rhythmic expression with a single peak and trough level was confirmed by RNA blot analysis for 3β-Hsd and Gabarapl1 genes exhibiting a peak in the light phase and Spot14, Hspa8, Hspa5, and Hsp84-1 genes showing a peak in the dark phase. On the other hand, mRNA levels for all of the three fibrinogen subunits, Aα, Bβ and γ, exhibited two peaks each in the light and dark phases in a synchronized manner. This two-peaked rhythmic pattern of fibrinogen genes as well as the single peaktrough pattern of other genes was diminished or almost completely lost in the liver of Clock mutant mice, suggesting that the two-peaked expression is also under the control of oscillation-generating genes. In constant darkness, the first peak of the expression rhythm of fibrinogen genes was almost intact, but the second peak disappeared. Therefore, although the first peak in the subjective day is a component of the innate circadian rhythm, the second peak seems to require light stimuli. Fasting in constant darkness caused shifts of time phases of the circadian rhythms. Protein levels of the fibrinogen subunits in whole blood also exhibited circadian rhythms. In the mouse and human loci of the fibrinogen gene cluster, a number of sequence elements resembling circadian transcription factor-binding sites were found. The fibrinogen gene locus provides a unique system for the study of two-peaked day/night rhythms of gene expression in a synchronized form. Genes expressed with day/night rhythms in the mouse liver were searched for by microarray analysis using an in-house array harboring mouse liver cDNAs. The rhythmic expression with a single peak and trough level was confirmed by RNA blot analysis for 3β-Hsd and Gabarapl1 genes exhibiting a peak in the light phase and Spot14, Hspa8, Hspa5, and Hsp84-1 genes showing a peak in the dark phase. On the other hand, mRNA levels for all of the three fibrinogen subunits, Aα, Bβ and γ, exhibited two peaks each in the light and dark phases in a synchronized manner. This two-peaked rhythmic pattern of fibrinogen genes as well as the single peaktrough pattern of other genes was diminished or almost completely lost in the liver of Clock mutant mice, suggesting that the two-peaked expression is also under the control of oscillation-generating genes. In constant darkness, the first peak of the expression rhythm of fibrinogen genes was almost intact, but the second peak disappeared. Therefore, although the first peak in the subjective day is a component of the innate circadian rhythm, the second peak seems to require light stimuli. Fasting in constant darkness caused shifts of time phases of the circadian rhythms. Protein levels of the fibrinogen subunits in whole blood also exhibited circadian rhythms. In the mouse and human loci of the fibrinogen gene cluster, a number of sequence elements resembling circadian transcription factor-binding sites were found. The fibrinogen gene locus provides a unique system for the study of two-peaked day/night rhythms of gene expression in a synchronized form. In mammals, intracellular oscillation generators for circadian rhythms reside in various peripheral organs such as liver as well as in the central pacemaker, the suprachiasmatic nucleus (SCN) 1The abbreviations used are: SCN, suprachiasmatic nucleus; ZT, Zeitgeber time; Hspa, heat shock 70-kDa protein; 3β-HSD, 3β-hydroxy-Δ5-C27-steroid oxidoreductase (dehydrogenase); Gabarapl1, GABA(A) receptor-associated protein-like 1; Hsp84-1, heat shock protein 84-kDa 1; Fg, fibrinogen; CT, circadian time; GRE, glucocorticoid response element; REV/ERBα, reverse strand gene of erbAα; DBP, albumin site D-binding protein. of the hypothalamus (1Ripperger J.A. Schibler U. Curr. Opin. Cell Biol. 2001; 13: 357-362Crossref PubMed Scopus (93) Google Scholar, 2Reppert S.M. Weaver D.R. Nature. 2002; 418: 935-941Crossref PubMed Scopus (3373) Google Scholar). The SCN and peripheral tissues seem to share common molecular mechanisms for generating the circadian oscillation. The widely accepted mechanism is autoregulatory feedback inhibition of period (Per) and cryptochrome (Cry) genes by their own protein products in antagonism with the positive transcription factor of the CLOCK-BMAL1 heterodimer (3Gekakis N. Staknis D. Nguyen H.B. Davis F.C. Wilsbacher L.D. King D.P. Takahashi J.S. Weitz C.J. Science. 1998; 280: 1564-1569Crossref PubMed Scopus (1555) Google Scholar, 4Jin X. Shearman L.P. Weaver D.R. Zylka M.J. de Vries G.J. Reppert S.M. Cell. 1999; 96: 57-68Abstract Full Text Full Text PDF PubMed Scopus (789) Google Scholar, 5Kume K. Zylka M.J. Sriram S. Shearman L.P. Weaver D.R. Jin X. Maywood E.S. Hastings M.H. Reppert S.M. Cell. 1999; 98: 193-205Abstract Full Text Full Text PDF PubMed Scopus (1308) Google Scholar, 6van 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. Buijs R. Bootsma D. Hoeijmakers J.H. Yasui A. Nature. 1999; 398: 627-630Crossref PubMed Scopus (1125) Google Scholar, 7Bunger 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. Cell. 2000; 103: 1009-1017Abstract Full Text Full Text PDF PubMed Scopus (1181) Google Scholar, 8Shearman L.P. Sriram S. Weaver D.R. Maywood E.S. Chaves I. Zheng B. Kume K. Lee C.C. van der Horst G.T. Hastings M.H. Reppert S.M. Science. 2000; 288: 1013-1019Crossref PubMed Scopus (1123) Google Scholar). The time phase of the SCN oscillator is adjusted by light stimuli everyday in the light-dark condition (1Ripperger J.A. Schibler U. Curr. Opin. Cell Biol. 2001; 13: 357-362Crossref PubMed Scopus (93) Google Scholar, 2Reppert S.M. Weaver D.R. Nature. 2002; 418: 935-941Crossref PubMed Scopus (3373) Google Scholar). The SCN then controls or affects the oscillators of peripheral cells in both relatively direct and indirect manners. The liver provides a well characterized example of these regulations. Although oscillation generators of liver cells are under the circadian control of glucocorticoids (9Balsalobre A. Brown S.A. Marcacci L. Tronche F. Kellendonk C. Reichardt H.M. Schütz G. Schibler U. Science. 2000; 289: 2344-2347Crossref PubMed Scopus (1384) Google Scholar, 10Le Minh N. Damiola F. Tronche F. Schütz G. Schibler U. EMBO J. 2001; 20: 7128-7136Crossref PubMed Scopus (389) Google Scholar), presumably through the hypothalamus-hypophysis-adrenal axis, time phases of the liver oscillators are more profoundly affected by feeding time (11Damiola F. Le Minh N. Preitner N. Kornmann B. Fleury-Olela F. Schibler U. Genes Dev. 2000; 14: 2950-2961Crossref PubMed Scopus (1747) Google Scholar, 12Hara R. Wan K. Wakamatsu H. Aida R. Moriya T. Akiyama M. Shibata S. Genes Cells. 2001; 6: 269-278Crossref PubMed Scopus (473) Google Scholar, 13Stokkan K.A. Yamazaki S. Tei H. Sakaki Y. Menaker M. Science. 2001; 291: 490-493Crossref PubMed Scopus (1368) Google Scholar). Artificial diurnal feeding of nocturnal rodents can completely uncouple the oscillation phase of the liver from that of the SCN. Seemingly, the SCN regulates the liver oscillators indirectly by controlling the sleep-awakeness phase that in turn determines the feeding time under the natural condition. Recently, hundreds of genes exhibiting circadian oscillation in their mRNA levels have been identified in the rodent liver (14Kornmann B. Preitner N. Rifat D. Fleury-Olela F. Schibler U. Nucleic Acids Res. 2001; 29: e51Crossref PubMed Scopus (117) Google Scholar, 15Akhtar R.A. Reddy A.B. Maywood E.S. Clayton J.D. King V.M. Smith A.G. Gant T.W. Hastings M.H. Kyriacou C.P. Curr. Biol. 2002; 12: 540-550Abstract Full Text Full Text PDF PubMed Scopus (648) Google Scholar, 16Kita Y. Shiozawa M. Jin W. Majewski R.R. Besharse J.C. Greene A.S. Jacob H.J. Pharmacogenetics. 2002; 12: 55-65Crossref PubMed Scopus (132) Google Scholar, 17Panda S. Antoch M.P. Miller B.H. Su A.I. Schook A.B. Straume M. Schultz P.G. Kay S.A. Takahashi J.S. Hogenesch J.B. Cell. 2002; 109: 307-320Abstract Full Text Full Text PDF PubMed Scopus (1874) Google Scholar, 18Storch K.F. Lipan O. Leykin I. Viswanathan N. Davis F.C. Wong W.H. Weitz C.J. Nature. 2002; 417: 78-83Crossref PubMed Scopus (1243) Google Scholar, 19Ueda H.R. Chen W. Adachi A. Wakamatsu H. Hayashi S. Takasugi T. Nagano M. Nakahama K. Suzuki Y. Sugano S. Iino M. Shigeyoshi Y. Hashimoto S. Nature. 2002; 418: 534-539Crossref PubMed Scopus (707) Google Scholar) by comprehensive gene expression analysis including the microarray analysis, showing that a number of liver functions are under the circadian control at the gene expression level. These genes seem to show mainly circadian rhythms with a single peak-trough a day, because most of the genes were identified by the algorithms examining the fitness to cosine wave patterns with ∼24 h of cycles in constant darkness. Here, in the course of cDNA microarray analysis and following RNA blot analysis, we found by chance that mRNAs for all of the three fibrinogen subunits, Aα, Bβ, and γ, exhibit peak levels twice a day under the light-dark condition in the mouse liver. Fibrinogen, a major component of blood coagulation, in the blood plasma has the subunit composition of (Aα)2(Bβ)2γ2 from which N-terminal peptides A and B are proteolytically removed by thrombin, yielding fibrin with the α2β2γ2 composition (20Mosesson M.W. Siebenlist K.R. Meh D.A. Ann. N. Y. Acad. Sci. 2001; 936: 11-30Crossref PubMed Scopus (461) Google Scholar). Apparently, corresponding with this stoichiometry of the subunit composition, the two-peaked rhythms of fibrinogen genes were synchronized. In constant darkness, the second peak of daily rhythms of fibrinogen genes disappeared, whereas the first peak was almost intact. In human (21Kant J.A. Fornace Jr., A.J. Saxe D. Simon M.I. McBride O.W. Crabtree G.R. Proc. Natl. Acad. Sci. U. S. A. 1985; 82: 2344-2348Crossref PubMed Scopus (242) Google Scholar) and rat (22Marino M.W. Fuller G.M. Elder F.F. Cytogenet. Cell Genet. 1986; 42: 36-41Crossref PubMed Google Scholar), the three fibrinogen genes were shown to be clustered on chromosome 4 (4q31) and chromosome 2 (2q31–34), respectively, which is also the case for mouse chromosome 3 (3 E3-F1) as confirmed by the genome sequence (23Mouse Genome Sequencing ConsortiumNature. 2002; 420: 520-562Crossref PubMed Scopus (5415) Google Scholar). Candidate sequence elements for the rhythmic regulation were mapped in the mouse and human loci of the fibrinogen gene cluster. Animals—Microarray analysis utilized 6-week-old male ddY mice (Takasugi Bioanimal, Saitama, Japan). Clock mutant C57BL/6J mice for analysis of rhythmicity were purchased from Jackson Laboratory (stock number 002923, Bar Harbor, ME) and interbred in Waseda University. Genotypes were determined by PCR as described previously (4Jin X. Shearman L.P. Weaver D.R. Zylka M.J. de Vries G.J. Reppert S.M. Cell. 1999; 96: 57-68Abstract Full Text Full Text PDF PubMed Scopus (789) Google Scholar). Mice were maintained on a light-dark cycle (12-h light, 12-h dark) at a room temperature of 23 °C and given food and water ad libitum. Lights-on time was assigned Zeitgeber time (ZT) 0, and then lights-off time was assigned ZT12. For experiments in constant darkness, C57BL/6N mice purchased from Charles River Japan Inc. (Yokohama, Japan) were housed under the light-dark condition for 2 weeks before transferring to constant darkness cDNA Microarray Analysis—A microarray was prepared essentially as described previously (24Yoshikawa T. Nagasugi Y. Azuma T. Kato M. Sugano S. Hashimoto K. Masuho Y. Muramatsu M. Seki N. Biochem. Biophys. Res. Commun. 2000; 275: 532-537Crossref PubMed Scopus (62) Google Scholar) with 2,304 mouse liver cDNA clones provided by K. Hashimoto (National Institute of Infectious Diseases, Tokyo, Japan) and S. Sugano (Institute of Medical Sciences, University of Tokyo, Tokyo, Japan). Total liver RNA was prepared from ddY mice at ZT6 and 18 by the acid-guanidine thiocyanate-phenol/chloroform extraction procedure (25Chomczynski P. Sacchi N. Anal. Biochem. 1987; 162: 156-159Crossref PubMed Scopus (63176) Google Scholar) and subjected to poly(A)+ RNA isolation using oligo(dT)-paramagnetic beads Dynabeads Oligo(dT)25 (Dynal, Oslo, Norway). 2 μg of poly(A)+ RNA and 4.5 μg of oligo(dT) in 15 μl of solution were heat-denatured at 70 °C for 10 min and immediately chilled on ice. The mixture was made up to 30 μl of the solution containing 50 mm Tris-HCl (pH 8.3), 75 mm KCl, 3 mm MgCl2, 10 mm DTT, 0.5 mm each of dATP, dCTP, and dGTP, 0.2 mm dTTP, 0.1 mm Cy3- or Cy5-dUTP, and 400 units of reverse transcriptase SuperScript II (Invitrogen). The reaction was allowed to proceed at 42 °C for 1 h. After alkaline degradation of template RNA, the Cy3- or Cy5-labeled cDNA was purified with Centricon-30 microconcentrators (Millipore, Bedford, MA). Hybridization was carried out in a solution containing 2 μg/μl yeast RNA, 2 μg/μl poly(A), 3.4× SSC (1× SSC consists of 0.15 m NaCl and 15 mm sodium citrate), and 0.3% SDS at 65 °C overnight under humidified condition. Washing was done twice for 5 min with 2× SSC, 0.1% SDS at room temperature and twice for 5 min with 0.2× SSC, 0.1% SDS at 40 °C followed by rinse with 0.2× SSC. The fluorescent images were scanned with a laser-scanning device (Scan-Array4000, GSI Lumonics, Bedford, MA). Northern Blot Analysis—Total liver RNA and sense strand cRNA amplified from poly(A)+ RNA were subjected to Northern analysis. The detailed procedure for cRNA amplification will be described elsewhere. Double-stranded cDNA synthesized from poly(A)+ RNA absorbed on the oligo(dT) beads was ligated with the T7 promoter adaptor and heat-denatured, liberating the sense strand cDNA, which was again converted to double-stranded form priming from the oligo(dT) linker, amplified with PCR using as primers known sequences of both ends, and then subjected to synthesis of sense strand cRNA with T7 RNA polymerase. RNAs were electrophoresed in denaturing formaldehyde-agarose (1%) gels, visualized by ethidium bromide staining to check integrity and equal loading, and then blotted onto nylon membranes. Digoxigenin-labeled RNA probes were synthesized using a transcription kit (Roche Diagnostics) from cDNAs subcloned in the plasmid pGEM-3Zf(+). Hybridization, washing, and chemiluminescent detection on x-ray films were done as recommended by Roche Diagnostics. Densitometric quantification was performed by using Personal Scanning Image (PDSI, Molecular Dynamics, Sunnyvale, CA). Western Blot Analysis—Whole blood was collected from the inferior vena cava of mice with heparinized syringes and tubes, and subjected to immunodetection of fibrinogen subunits essentially as described previously (26Heckel J.L. Sandgren E.P. Degen J.L. Palmiter R.D. Brinster R.L. Cell. 1990; 62: 447-456Abstract Full Text PDF PubMed Scopus (180) Google Scholar). The blood was serially diluted with 19 and 5.25 volumes of 70 mm Tris-HCl buffer (pH 6.8) containing 3% SDS, 5% 2-mercaptoethanol, 11.2% glycerol, and 0.02% bromphenol blue. The diluted samples containing 0.08 μl of whole blood in 20 μl of solution were subjected to SDS-PAGE with 10% acrylamide gel, and proteins were electrotransferred to nitrocellulose membranes. Immunodetection was performed using a goat IgG against mouse fibrinogen (1:1,000 dilution, Nordic Immunological Laboratories, Tilburg, The Netherlands) as a primary antibody, a peroxidase-conjugated rabbit IgG F(ab′)2 against the goat IgG (1:5,000 dilution, Wako Pure Chemical Industries, Osaka, Japan) as a secondary antibody and ImmunoStar Reagents (Wako, Japan) for chemiluminescence detection. Signals detected on x-ray films were densitometrically quantified. Data Base Search—cDNA sequences deposited in IMAGE were accessed through Search GenBank™-Today Data Base using DBGET (www.genome.ad.jp/dbget-bin/www_bfind?genbank-today), queried in NCBI BLAST (www.ncbi.nlm.nih.gov:80/BLAST/), and assigned to specific genes with NCBI UniGene (www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=unigene) and NCBI LocusLink (www.ncbi.nlm.nih.gov/LocusLink/). Mouse and human genome sequences were from Mouse Genome Server (www.ensembl.org/Mus_musculus/) and Human Genome Server (www.ensembl.org/Homo_sapiens/), respectively, in Ensembl (www.ensembl.org/). cDNA Microarray Analysis of Day/Night Gene Expression in the Mouse Liver—Microarray analysis was carried out to detect differentially expressed genes between light (ZT6) and dark (ZT18) phases. We used an in-house microarray (24Yoshikawa T. Nagasugi Y. Azuma T. Kato M. Sugano S. Hashimoto K. Masuho Y. Muramatsu M. Seki N. Biochem. Biophys. Res. Commun. 2000; 275: 532-537Crossref PubMed Scopus (62) Google Scholar) containing ∼2,300 mouse liver cDNA clones. Poly(A)+ RNAs derived from the liver at ZT6 and ZT18 in duplicate were subjected to Cy3 and Cy5 labeling, respectively, coupled with cDNA synthesis. Both labeled cDNAs were mixed in an equal amount and hybridized with a microarray. Labeling vice versa with Cy3 and Cy5 was also done to correct a possible difference in incorporation efficiency of these dyes. Tables I and II represent lists for genes that showed the higher mRNA levels at ZT6 and ZT18, respectively, at least 1.5-fold in all four experiments of the microarray analysis. Remarkably, genes for all of the three fibrinogen subunits, Aα, Bβ and γ, showed the higher expression at ZT6 (Table I). It is also noteworthy that a number of genes possibly involved in gene expression ranging from transcription to protein transport showed the higher expression at ZT18 (Table I). Spot14 is a putative transcriptional regulator (27Brown S.B. Maloney M. Kinlaw W.B. J. Biol. Chem. 1997; 272: 2163-2166Abstract Full Text Full Text PDF PubMed Scopus (83) Google Scholar, 28Compe E. De Sousa G. François K. Roche R. Rahmani R. Torresani J. Raymondjean M. Planells R. Biochem. J. 2001; 358: 175-183Crossref PubMed Google Scholar), and leucine zipper protein 1 is a basic protein localized in the nucleus (29Sun D.S. Chang A.C. Jenkins N.A. Gilbert D.J. Copeland N.G. Chang N.C. Genomics. 1996; 36: 54-62Crossref PubMed Scopus (18) Google Scholar). Heat shock 70-kDa protein 8 (Hspa8), heat shock 70-kDa protein 5 (Hspa5) (78-kDa glucose-regulated protein), and heat shock protein 84-kDa 1 (Hsp84-1) are molecular chaperones (30Frydman J. Annu. Rev. Biochem. 2001; 70: 603-647Crossref PubMed Scopus (934) Google Scholar), and peroxisome biogenesis factor 7 is involved in posttranslational protein transport into the organelle (31Distel B. Erdmann R. Gould S.J. Blobel G. Crane D.I. Cregg J.M. Dodt G. Fujiki Y. Goodman J.M. Just W.W. Kiel J.A. Kunau W.H. Lazarow P.B. Mannaerts G.P. Moser H.W. Osumi T. Rachubinski R.A. Roscher A. Subramani S. Tabak H.F. Tsukamoto T. Valle D. van der Klei I. van Veldhoven P.P. Veenhuis M. J. Cell Biol. 1996; 135: 1-3Crossref PubMed Scopus (313) Google Scholar, 32Braverman N. Steel G. Lin P. Moser A. Moser H. Valle D. Genomics. 2000; 63: 181-192Crossref PubMed Scopus (37) Google Scholar). Seemingly, these genes are involved in synthesis and delivery of proteins in a dark phase-specific manner.Table IGenes exhibiting the higher expression in the light phase (ZT6)IMAGE clone no.ZT6/ZT18 ratioGene nameSymbolFunctionUniGene clusterLocus-Link IDNCBI reference sequenceMapPair 1Pair 2AverageCyto-geneticGeneticCy5/Cy3Cy3/Cy5Cy5/Cy3Cy3/Cy5cM acM, centimorgan.14504392.432.993.072.482.743β-hydroxy-Δ5-C27-steroid oxidoreductase3BETA-HSDBile acid synthesisMm.157103101502XM_1337657 F314515132.262.602.792.512.54GABA(A) receptor-associated protein-like 1Gabarapl1Cytoskeletal protein bindingMm.1463857436XM_1328496 F314311633.482.572.071.532.41Fibrinogen, Bβ polypeptideFgbBlood coagulationMm.30063110135XM_1309603 E33 48.214825652.662.521.921.782.22Claudin 1Cldn1Integral membrane proteinMm.3366912737NM_01667416 B114315862.522.681.591.822.15Fibrinogen, Aα polypeptideFgaBlood coagulationMm.8879314161XM_1309313 F114311851.872.761.941.612.05Fibrinogen, γ polypeptideFggBlood coagulationMm.1642299571XM_1240633 E314310562.212.041.871.851.99Cysteine dioxygenase 1, cytosolicCdo1Amino acid metabolismMm.2999612583XM_12321718 C18 23.0a cM, centimorgan. Open table in a new tab Table IIGenes exhibiting the higher expression in the dark phase (ZT18)IMAGE clone no.ZT18/ZT6 ratioGene nameSymbolFunctionUniGene clusterLocus-Link IDNCBI reference sequenceMapPair 1Pair 2AverageCyto-geneticGeneticCy3/Cy5Cy5/Cy3Cy3/Cy5Cy5/Cy3cM acM, centimorgan.14997245.985.599.198.827.39Spot14Spot14Transcription regulationMm.2858521835NM_0093817 E114317065.793.935.083.564.59Heat shock 70-kDacM, centimorgan. protein 8Hspa8ChaperoneMm.19755115481XM_1347819 A5.19 24.014512344.084.304.353.213.98Heat shock 70-kDacM, centimorgan. protein 5 (78-kD GRP)Hspa5ER stress responseMm.91814828XM_1237822 B2 22.514508742.762.052.602.622.51Heat shock protein 84-kDa 1Hsp84-1ChaperoneMm.218015516XM_12468817 B317 15.014996502.521.823.042.232.40Flavin containing monooxygenase 5Fmo5Electron transportMm.166814263NM_0102323 F2.218879812.471.832.062.032.10Major urinary protein 1Mup1Pheromone bindingMm.15789317840XM_13557044 27.814516252.821.602.091.852.09Nicotinamide N-methyltransferaseNnmtNiacin metabolismMm.836218113XM_1347989 A539 29.018860072.201.562.192.062.00Leucine zipper protein 1Luzp1Leucine zipper proteinMm.9265917024NM_0244524 D34 66.418872701.701.662.511.581.87Peroxisome biogenesis factor 7Pex7Intracellular protein transportMm.244018634NM_00882210 A3a cM, centimorgan. Open table in a new tab To confirm the validity of the microarray analysis, we performed Northern analysis for six genes, i.e. two and four genes that showed the higher expression at ZT6 and ZT18, respectively, more than 2-fold in all four experiments. Total liver RNAs at ZT6 and ZT18 in duplicate were subjected to the blot analysis (Fig. 1A). Concordant with the results of the microarray analysis, mRNA levels for 3β-hydroxy-Δ5-C27-steroid oxidoreductase (3β-HSD) (33Schwarz M. Wright A.C. Davis D.L. Nazer H. Bjorkhem I. Russell D.W. J. Clin. Invest. 2000; 106: 1175-1184Crossref PubMed Scopus (84) Google Scholar) and GABA(A) receptor-associated protein-like 1 (Gabarapl1) (34Xin Y. Yu L. Chen Z. Zheng L. Fu Q. Jiang J. Zhang P. Gong R. Zhao S. Genomics. 2001; 74: 408-413Crossref PubMed Scopus (83) Google Scholar) were higher at ZT6 than ZT18 and those for Spot14, Hspa8, Hspa5, and Hsp84-1 were higher at ZT18 than ZT6. We also performed Northern analysis by using poly(A)+ RNA-derived sense strand cRNA (Fig. 1B) to exclude the possibility that the results obtained with total RNA may reflect daily fluctuation of poly(A)+ RNA contents in total RNA. Again, changes in the gene expression levels detected with the cRNA mixtures were concordant with the results of the microarray analysis for both gene groups each exhibiting the higher expression at ZT6 or ZT18. We concluded that our microarray system successfully revealed genes differentially expressed between light and dark phases. Genes Exhibiting Rhythmic Expression with a Single Peak-Trough—Rhythmic patterns of expression of the six genes were examined by Northern analysis with total RNA extracted from the liver of wild-type mice at 4-h intervals in a light-dark cycle. In parallel, we also investigated the mice with Clock mutation that yields a splice variant of a dominant-negative type (35King 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. Turek F.W. Takahashi J.S. Cell. 1997; 89: 641-653Abstract Full Text Full Text PDF PubMed Scopus (1146) Google Scholar) to see whether the rhythmic expression of the six genes are under the control of oscillation-generating genes. As shown in Fig. 2A, mRNA levels for 3β-HSD and Gabarapl1 in the wild-type mice exhibited day/night rhythm with a peak in the light phase and a trough in the dark phase. In the Clock mutant mice, the rhythmic oscillations of these genes were apparently attenuated, indicating that CLOCK plays a role in the regulation of 3β-Hsd and Gabarapl1 genes. Fig. 2B shows the results of similar analysis on Spot14, Hspa8, Hspa5, and Hsp84-1 genes with the higher expression in the dark phase. In the wild-type mice, each mRNA level for these genes showed a peak in the dark phase and a trough in the light phase. On the other hand, the rhythmic oscillations of the four genes were attenuated or almost completely lost in Clock mutant mice. Taken together with the results of Fig. 2A, both gene groups, each displaying a peak expression in the light or dark phase, seem to be under the control of the Clock gene. Two-peaked Day/Night Rhythms in Expression of Fibrinogen Genes—We also examined daily changes in expression of genes for fibrinogen subunits by Northern analysis (Fig. 3). Interestingly, in wild-type mice, mRNA levels for all three fibrinogen subunits, Aα, Bβ, and γ, peaked twice at ZT7 and 19 in a day. The time phase and oscillation amplitude of the two-peaked rhythm were synchronized among the three genes. Since fibrinogen protein in the blood plasma consists of three pairs of each subunit leading to the composition (Aα)2(Bβ)2γ2 (20Mosesson M.W. Siebenlist K.R. Meh D.A. Ann. N. Y. Acad. Sci. 2001; 936: 11-30Crossref PubMed Scopus (461) Google Scholar), synchronization in expression of the subunit genes is suited for maintenance of this stoichiometry. In Clock mutant mice, the rhythmicity appeared to be shifted to the weak single peak-trough pattern for the Aα gene or almost completely lost for the Bβ and γ genes. Therefore, the two-peaked expression of fibrinogen genes is also likely to be under the control of the Clock gene. To examine effects of the light-dark cycle and food intake on the daily rhythms of expression of fibrinogen genes, mice were kept in constant darkness and half of them were fasted (Fig. 4). Mice transferred to constant darkness on day 1 were subjected to liver excision at 4-h intervals from circadian time (CT) 3 to CT23 on day 2 (Fig. 4A). In Northern analysis for the fibrinogen subunit mRNAs (Fig. 4B, left panels), the second peak observed in the nighttime under the light-dark condition was not detected in the subjective night under constant darkness. On the other hand, the first peak was almost intact with some prolongation of the peak phase. Therefore, while the first peak in the subjective day is caused mainly by the innate circadian mechanism, the second peak seems to be generated by light stimuli (i.e. lights-on, continuous irradiation, and/or lights-off) most probably by lights-off at least as a component that affects the time phase of the second peak. Effects of food depletion were also examined, because it is well known that feeding time profoundly affects the time phases of the liver oscillators (11Damiola F. Le Minh N. Preitner N. Kornmann B. Fleury-Olela F. Schibler U. Genes Dev. 2000; 14: 2950-2961Crossref PubMed Scopus (1747) Google Scholar, 12Hara R. Wan K. Wakamatsu H. Aida R. Moriya T. Akiyama M. Shibata S. Genes Cells. 2001; 6: 269-278Crossref PubMed Scopus (473) Google Scholar, 13Stokkan K.A. Yamazaki S. Tei H. Sakaki Y. Menaker M. Science. 2001; 291: 490-493Crossref PubMed Scopus (1368) Google Scholar). Foods were depleted at CT11 on day 1 just before beginning of the feeding period of nocturnal mice in the subjective night (Fig. 4A). Fasting caused shifts of time phases in the expression rhythms of fibrinogen subunit genes (Fig. 4B, right panels). Directions of the shifts remain to be determined experimentally, but we assume that forward shifts took place because fasting brings about forward shifts of the rhythms of mPer1 and mPer2 mRNA levels in the liver. 2K. Horikawa, Y. Minami, M. Iijima, R. Aida, M. Akiyama, and S. Shibata, manuscript in preparation. Synchronization of the time phase among the subunit mRNAs seemed to be loosened by the fasting, the reason being remains to be clarified. Rhythms of protein levels for the fibrinogen subunits in constant darkness were examined by Western analysis with whole blood (Fig. 4C). When mice were fed ad libitum (left panels), the three subunits exhibited synchronized circadian rhythms with an apparent peak at CT7 and possibly a low peak at CT15. Compared with the mRNA rhythms (Fig. 4B, left panels), this is rather an unexpected result. Direct reflection of mRNA levels to protein levels would have resulted in mimicking mRNA patterns by protein patterns with the lag time of several hours. Seemingly, circadian rhythms of fibrinogen protein levels are controlled not only by mRNA levels but also by other mechanisms such as regulations in the steps of translation, secretion, and protein degradation. Again, fasting appeared to cause shifts of time phases of fibrinogen protein rhythms, although individual variations of protein levels were relatively large under the fasting. In human, it was demonstrated that plasma fibrinogen concentrations exhibit a day/night rhythm (36Petralito A. Mangiafico R.A. Gibiino S. Cuffari M.A. Miano M.F. Fiore C.E. Chronobiologia. 1982; 9: 195-201PubMed Google Scholar, 37Haus E. Cusulos M. Sackett-Lundeen L. Swoyer J. Chronobiol. Int. 1990; 7: 203-216Crossref PubMed Scopus (104) Google Scholar, 38Kanabrocki E.L. Sothern R.B. Messmore H.L. Roitman-Johnson B. McCormick J.B. Dawson S. Bremner F.W. Third J.L. Nemchausky B.A. Shirazi P. Scheving L.E. Clin. Appl. Thromb. Hemost. 1999; 5: 37-42Crossref PubMed Scopus (42) Google Scholar). Whereas the rhythm has been approximated by the cosine curve with a single peak in the early morning (38Kanabrocki E.L. Sothern R.B. Messmore H.L. Roitman-Johnson B. McCormick J.B. Dawson S. Bremner F.W. Third J.L. Nemchausky B.A. Shirazi P. Scheving L.E. Clin. Appl. Thromb. Hemost. 1999; 5: 37-42Crossref PubMed Scopus (42) Google Scholar), reconsideration of the original data of Kanabrocki et al. (38Kanabrocki E.L. Sothern R.B. Messmore H.L. Roitman-Johnson B. McCormick J.B. Dawson S. Bremner F.W. Third J.L. Nemchausky B.A. Shirazi P. Scheving L.E. Clin. Appl. Thromb. Hemost. 1999; 5: 37-42Crossref PubMed Scopus (42) Google Scholar) suggests the presence of the second peak or a shoulder in the evening. Therefore, human fibrinogen seems to be set to increase rapidly in the morning and persist around the daytime, i.e. a physically active period in human. The peak in the early morning has been implicated in the incidence of arterial ischemic diseases such as myocardial infarction (38Kanabrocki E.L. Sothern R.B. Messmore H.L. Roitman-Johnson B. McCormick J.B. Dawson S. Bremner F.W. Third J.L. Nemchausky B.A. Shirazi P. Scheving L.E. Clin. Appl. Thromb. Hemost. 1999; 5: 37-42Crossref PubMed Scopus (42) Google Scholar). Putative Regulatory Elements for Rhythmic Expression of the Fibrinogen Gene Cluster—The two-peaked rhythmicity suggests that at least two or possibly more transcription factors are involved in daily regulation of the fibrinogen subunit genes. On the other hand, synchronization in expression of subunit genes implies that they share common regulatory mechanisms. We searched for putative cis-regulatory elements interacting with circadian transcription factors in the fibrinogen gene cluster (Fig. 5). Recent determination of the mouse genome sequence (23Mouse Genome Sequencing ConsortiumNature. 2002; 420: 520-562Crossref PubMed Scopus (5415) Google Scholar) confirmed the presence of the fibrinogen gene cluster, which was predictable from the cytogenetic map (Table I) and from previous reports on homologous clusters of human (21Kant J.A. Fornace Jr., A.J. Saxe D. Simon M.I. McBride O.W. Crabtree G.R. Proc. Natl. Acad. Sci. U. S. A. 1985; 82: 2344-2348Crossref PubMed Scopus (242) Google Scholar) and rat (22Marino M.W. Fuller G.M. Elder F.F. Cytogenet. Cell Genet. 1986; 42: 36-41Crossref PubMed Google Scholar). The searched sequences were as follows: the E-box element CACGTG that is recognized by CLOCK/BMAL1 (3Gekakis N. Staknis D. Nguyen H.B. Davis F.C. Wilsbacher L.D. King D.P. Takahashi J.S. Weitz C.J. Science. 1998; 280: 1564-1569Crossref PubMed Scopus (1555) Google Scholar, 39Darlington T.K. Wager-Smith K. Ceriani M.F. Staknis D. Gekakis N. Steeves T.D. Weitz C.J. Takahashi J.S. Kay S.A. Science. 1998; 280: 1599-1603Crossref PubMed Scopus (694) Google Scholar) and NPAS2/BMAL1 (40Rutter J. Reick M. Wu L.C. McKnight S.L. Science. 2001; 293: 510-514Crossref PubMed Scopus (770) Google Scholar) and that is also the target of repressors such as PER/CRY (5Kume K. Zylka M.J. Sriram S. Shearman L.P. Weaver D.R. Jin X. Maywood E.S. Hastings M.H. Reppert S.M. Cell. 1999; 98: 193-205Abstract Full Text Full Text PDF PubMed Scopus (1308) Google Scholar, 8Shearman L.P. Sriram S. Weaver D.R. Maywood E.S. Chaves I. Zheng B. Kume K. Lee C.C. van der Horst G.T. Hastings M.H. Reppert S.M. Science. 2000; 288: 1013-1019Crossref PubMed Scopus (1123) Google Scholar) and recently characterized Dec1 and Dec2 (41Honma S. Kawamoto T. Takagi Y. Fujimoto K. Sato F. Noshiro M. Kato Y. Honma K. Nature. 2002; 419: 841-844Crossref PubMed Scopus (519) Google Scholar); the REV-ERBα site WAWNTRGGTCA (W = A or T, R = A or G) that is recognized by members of nuclear orphan receptor REV-ERB and retinoic acid-related orphan receptor families and that is responsible for circadian regulation (19Ueda H.R. Chen W. Adachi A. Wakamatsu H. Hayashi S. Takasugi T. Nagano M. Nakahama K. Suzuki Y. Sugano S. Iino M. Shigeyoshi Y. Hashimoto S. Nature. 2002; 418: 534-539Crossref PubMed Scopus (707) Google Scholar) by REV-ERBα as a well characterized member (42Preitner N. Damiola F. Luis-Lopez-Molina Z akany J. Duboule D. Albrecht U. Schibler U. Cell. 2002; 110: 251-260Abstract Full Text Full Text PDF PubMed Scopus (1675) Google Scholar); the DBP site RTTAYGTAAY (R = AorG, Y = C or T) that is the binding sequence of circadian transcription factors DBP, thyrotroph embryonic factor, and hepatocyte leukemia factor of the PAR-ZIP (proline- and acidic amino acid-rich region-leucine zipper) family (43Fonjallaz P. Ossipow V. Wanner G. Schibler U. EMBO J. 1996; 15: 351-362Crossref PubMed Scopus (124) Google Scholar) and its variant E4BP4 (44Mitsui S. Yamaguchi S. Matsuo T. Ishida Y. Okamura H. Genes Dev. 2001; 15: 995-1006Crossref PubMed Scopus (300) Google Scholar); and the glucocorticoid response element (GRE) GGTACANNNTGTTCT (45Evans R.M. Science. 1988; 240: 889-895Crossref PubMed Scopus (6323) Google Scholar) that is the recognition sequence of the nuclear receptor bound and activated by glucocorticoids, a major humoral regulator of the liver clock (9Balsalobre A. Brown S.A. Marcacci L. Tronche F. Kellendonk C. Reichardt H.M. Schütz G. Schibler U. Science. 2000; 289: 2344-2347Crossref PubMed Scopus (1384) Google Scholar, 10Le Minh N. Damiola F. Tronche F. Schütz G. Schibler U. EMBO J. 2001; 20: 7128-7136Crossref PubMed Scopus (389) Google Scholar). Completely matched sequences for the E-box and one mismatch-allowed sequence for the other sites are marked on mouse (Fig. 5A) and human (Fig. 5B) fibrinogen gene cluster loci with 10-kb flanking regions. A number of candidate sequences for each site were found. Exceptionally, only one GRE-like sequence was detected in both mouse and human, and its location is relatively similar between the two species, i.e. around the 3′-terminal region of the fibrinogen Bβ gene. Therefore, this putative GRE may be functional and conservatory. Previous studies have proposed the presence of GREs for rat Aα and Bβ genes but have not identified them (46Fuller G.M. Zhang Z. Ann. N. Y. Acad. Sci. 2001; 936: 469-479Crossref PubMed Scopus (91) Google Scholar). In the 5′-flanking region of the γ gene, sequences each resembling the E-box, REV-ERBα site, or DBP site seem to be arranged similarly between mouse and human and may play regulatory roles common to both species. On the other hand, frequencies of the three elements in the whole locus are rather different between the two species and each frequency in the mouse versus human locus is as follows (the number within parentheses shows the frequency of completely matched sequences): the E-box 9 (9Balsalobre A. Brown S.A. Marcacci L. Tronche F. Kellendonk C. Reichardt H.M. Schütz G. Schibler U. Science. 2000; 289: 2344-2347Crossref PubMed Scopus (1384) Google Scholar) versus 3 (3Gekakis N. Staknis D. Nguyen H.B. Davis F.C. Wilsbacher L.D. King D.P. Takahashi J.S. Weitz C.J. Science. 1998; 280: 1564-1569Crossref PubMed Scopus (1555) Google Scholar); the REV-ERBα site 25 (4Jin X. Shearman L.P. Weaver D.R. Zylka M.J. de Vries G.J. Reppert S.M. Cell. 1999; 96: 57-68Abstract Full Text Full Text PDF PubMed Scopus (789) Google Scholar) versus 41 (2Reppert S.M. Weaver D.R. Nature. 2002; 418: 935-941Crossref PubMed Scopus (3373) Google Scholar); and the DBP site 16 (0) versus 44 (2Reppert S.M. Weaver D.R. Nature. 2002; 418: 935-941Crossref PubMed Scopus (3373) Google Scholar). It is tempting to speculate that relatively high frequencies of the E-box in mouse and the REV-ERBα and DBP sites in human are implicated in species-specific regulations in mouse and human adapted to nocturnal and diurnal activity, respectively. It will be extremely interesting to investigate the roles of these putative cis-elements in two-peaked synchronization in the expression of the fibrinogen gene cluster that is under the control of both innate circadian mechanism and seemingly light-responsive mechanism. We are grateful to K. Hashimoto and S. Sugano for providing us cDNA clones for the microarray. We also thank T. Hiwasa and our colleagues for suggestions, help, and discussions.

Referência(s)