Portal de Periódicos da CAPES

Increased Hepatic Forkhead Box M1B (FoxM1B) Levels in Old-aged Mice Stimulated Liver Regeneration through Diminished p27Kip1 Protein Levels and Increased Cdc25B Expression

2002; Elsevier BV; Volume: 277; Issue: 46 Linguagem: Inglês

10.1074/jbc.m207510200

1083-351X

Xinhe Wang, Katherine Krupczak-Hollis, Yongjun Tan, Margaret B. Dennewitz, Guy R. Adami, Robert H. Costa,

Hippo pathway signaling and YAP/TAZ

Recent liver regeneration studies indicate that maintaining hepatic Forkhead Box M1B (FoxM1B) expression in 12-month-old (old-aged) Transthyretin-FoxM1B transgenic mice increases hepatocyte proliferation and expression of cell cycle regulatory genes. Because these transgenic CD-1 mice maintain FoxM1B levels during the aging process, we conducted the current study to determine whether adenovirus delivery of the FoxM1B gene (AdFoxM1B) is sufficient to stimulate liver regeneration in old-aged Balb/c mice. Here we show that AdFoxM1B infection of old-aged mice caused a significant increase in FoxM1B expression, hepatocyte DNA replication, and mitosis following partial hepatectomy. This stimulation in hepatocyte S-phase progression was associated with diminished protein expression and perinuclear localization of cyclin-dependent kinase (Cdk) inhibitor p27Kip1 (p27) protein following partial hepatectomy. In contrast, old-aged mice infected with control virus displayed high hepatocyte levels of p27 protein, which had been localized to the nucleus prior to S-phase. Furthermore, we found that restoring FoxM1B expression did not influence p27 mRNA levels, and this new finding implicates FoxM1B in regulation of p27 protein levels. Likewise, AdFoxM1B-infected regenerating livers displayed elevated S-phase levels of Cdk2 kinase activity compared with old-aged mice infected with control virus. Furthermore, restoring FoxM1B expression in old-aged mice caused elevated levels of Cyclin B1, Cyclin B2, Cdc25B, Cdk1, and p55CDC mRNA as well as stimulating Cdc25B nuclear localization during liver regeneration, all of which are required for mitosis. These studies indicated that an acute delivery of the FoxM1B gene in old-aged mice is sufficient to re-establish proliferation of regenerating hepatocytes, suggesting that FoxM1B can be used for therapeutic intervention to alleviate the reduction in cellular proliferation observed in the elderly. Recent liver regeneration studies indicate that maintaining hepatic Forkhead Box M1B (FoxM1B) expression in 12-month-old (old-aged) Transthyretin-FoxM1B transgenic mice increases hepatocyte proliferation and expression of cell cycle regulatory genes. Because these transgenic CD-1 mice maintain FoxM1B levels during the aging process, we conducted the current study to determine whether adenovirus delivery of the FoxM1B gene (AdFoxM1B) is sufficient to stimulate liver regeneration in old-aged Balb/c mice. Here we show that AdFoxM1B infection of old-aged mice caused a significant increase in FoxM1B expression, hepatocyte DNA replication, and mitosis following partial hepatectomy. This stimulation in hepatocyte S-phase progression was associated with diminished protein expression and perinuclear localization of cyclin-dependent kinase (Cdk) inhibitor p27Kip1 (p27) protein following partial hepatectomy. In contrast, old-aged mice infected with control virus displayed high hepatocyte levels of p27 protein, which had been localized to the nucleus prior to S-phase. Furthermore, we found that restoring FoxM1B expression did not influence p27 mRNA levels, and this new finding implicates FoxM1B in regulation of p27 protein levels. Likewise, AdFoxM1B-infected regenerating livers displayed elevated S-phase levels of Cdk2 kinase activity compared with old-aged mice infected with control virus. Furthermore, restoring FoxM1B expression in old-aged mice caused elevated levels of Cyclin B1, Cyclin B2, Cdc25B, Cdk1, and p55CDC mRNA as well as stimulating Cdc25B nuclear localization during liver regeneration, all of which are required for mitosis. These studies indicated that an acute delivery of the FoxM1B gene in old-aged mice is sufficient to re-establish proliferation of regenerating hepatocytes, suggesting that FoxM1B can be used for therapeutic intervention to alleviate the reduction in cellular proliferation observed in the elderly. Liver regeneration induced by two thirds partial hepatectomy (PHx) 1The abbreviations used are: PHx, partial hepatectomy; Fox, Forkhead Box; FoxM1B, Forkhead Box M1B; AdFoxM1B, adenovirus-expressing FoxM1B; AdEmpty, replication-defective adenovirus control lacking the E1 region; MI, mock-infected; TG, transgenic; APC, anaphase-promoting complex; Cdk, cyclin-dependent kinase; TTR, Transthyretin; p27Kip1, Cdk inhibitor; RB, retinoblastoma; PBS, phosphate-buffered saline; BrdUrd, 5-bromo-2′-deoxyuridine results in synchronous induction of hepatocyte proliferation through collaborative stimulation by growth factors and the interleukin 6 (IL-6) cytokine (1Cressman D.E. Greenbaum L.E. DeAngelis R.A. Ciliberto G. Furth E.E. Poli V. Taub R. Science. 1996; 274: 1379-1383Crossref PubMed Scopus (1324) Google Scholar, 2Diehl A.M. Immunol. Rev. 2000; 174: 160-171Crossref PubMed Scopus (268) Google Scholar, 3Michalopoulos G.K. DeFrances M.C. Science. 1997; 276: 60-66Crossref PubMed Scopus (2917) Google Scholar, 4Taub R. Greenbaum L.E. Peng Y. Semin. Liver Dis. 1999; 19: 117-127Crossref PubMed Scopus (133) Google Scholar, 5Yamada Y. Kirillova I. Peschon J.J. Fausto N. Proc. Natl. Acad. Sci. U. S. A. 1997; 94: 1441-1446Crossref PubMed Scopus (842) Google Scholar). Cell cycle progression is regulated by cyclin expression and degradation as well as by cyclin-mediated activation of cyclin-dependent kinases (Cdks). Formation of the Cyclin E-Cdk2 and Cyclin A-Cdk2 complexes is critical for S-phase entry and progression because they phosphorylate the retinoblastoma (RB) protein, which releases bound E2F transcription factor and allows it to stimulate expression of proliferation-specific target genes (6Harbour J.W. Dean D.C. Genes Dev. 2000; 14: 2393-2409Crossref PubMed Scopus (963) Google Scholar). Likewise, Cyclin B proteins associate with Cdk1 (Cdc2) to mediate cell cycle progression from the G2 phase into mitosis (7Zachariae W. Nasmyth K. Genes Dev. 1999; 13: 2039-2058Crossref PubMed Scopus (576) Google Scholar). Furthermore, Cdk activity is stimulated through dephosphorylation by the Cdc25A, Cdc25B, and Cdc25C phosphatase proteins (8Nilsson I. Hoffmann I. Prog. Cell Cycle Res. 2000; 4: 107-114Crossref PubMed Scopus (384) Google Scholar, 9Sebastian B. Kakizuka A. Hunter T. Proc. Natl. Acad. Sci. U. S. A. 1993; 90: 3521-3524Crossref PubMed Scopus (213) Google Scholar, 10Trembley J.H. Ebbert J.O. Kren B.T. Steer C.J. Cell Growth & Differ. 1996; 7: 903-916PubMed Google Scholar) and is negatively regulated by high levels of Cdk inhibitor p21Cip1, p27Kip1, and p16INK4A proteins (11Sherr C.J. Roberts J.M. Genes Dev. 1999; 13: 1501-1512Crossref PubMed Scopus (5158) Google Scholar). Finally, increased p55CDC protein levels are required to regulate the ubiquitin-ligase anaphase-promoting complex (APC)-mediated degradation of the Cyclin B proteins, whose elimination is required for the completion of mitosis (7Zachariae W. Nasmyth K. Genes Dev. 1999; 13: 2039-2058Crossref PubMed Scopus (576) Google Scholar). The Forkhead Box (Fox) transcription factors are an extensive family of transcription factors, consisting of more than 50 mammalian proteins (12Kaestner K.H. Knochel W. Martinez D.E. Genes Dev. 2000; 14: 142-146PubMed Google Scholar), which shares homology in the winged helix DNA-binding domain (13Clark K.L. Halay E.D. Lai E. Burley S.K. Nature. 1993; 364: 412-420Crossref PubMed Scopus (1097) Google Scholar). Its members play important roles in regulating cellular proliferation, differentiation, and metabolic homeostasis (14Costa R.H. Kalinichenko V.V. Lim L. Am. J. Physiol. Lung Cell Mol. Physiol. 2001; 280: L823-L838Crossref PubMed Google Scholar, 15Duncan S.A. Dev. Dyn. 2000; 219: 131-142Crossref PubMed Scopus (92) Google Scholar, 16Kaestner K.H. Trends Endocrinol. Metab. 2000; 11: 281-285Abstract Full Text Full Text PDF PubMed Scopus (151) Google Scholar, 17Kaufmann E. Knochel W. Mech. Dev. 1996; 57: 3-20Crossref PubMed Scopus (577) Google Scholar, 18Wang X. Quail E. Hung N.-J. Tan Y., Ye, H. Costa R.H. Proc. Natl. Acad. Sci. U. S. A. 2001; 98: 11468-11473Crossref PubMed Scopus (176) Google Scholar, 19Ye H. Holterman A. Yoo K.W. Franks R.R. Costa R.H. Mol. Cell. Biol. 1999; 19: 8570-8580Crossref PubMed Scopus (167) Google Scholar, 20Zaret K. Dev. Biol. 1999; 209: 1-10Crossref PubMed Scopus (185) Google Scholar). In regenerating liver, increased hepatic expression of FoxM1B levels occurs at the G1/S transition of the cell cycle (32 h post-PHx), and its levels remain elevated throughout the period of proliferation (19Ye H. Holterman A. Yoo K.W. Franks R.R. Costa R.H. Mol. Cell. Biol. 1999; 19: 8570-8580Crossref PubMed Scopus (167) Google Scholar, 21Ye H. Kelly T.F. Samadani U. Lim L. Rubio S. Overdier D.G. Roebuck K.A. Costa R.H. Mol. Cell. Biol. 1997; 17: 1626-1641Crossref PubMed Scopus (315) Google Scholar). To examine the role of FoxM1B in regulating cellular proliferation, we developed transgenic (TG) CD-1 mice in which the −3-kb Transthyretin (TTR) promoter was used to drive hepatocyte expression of the human FoxM1B (HFH-11B) cDNA (19Ye H. Holterman A. Yoo K.W. Franks R.R. Costa R.H. Mol. Cell. Biol. 1999; 19: 8570-8580Crossref PubMed Scopus (167) Google Scholar). Using these TG mice for liver regeneration studies, we found that premature FoxM1B expression caused an 8-h acceleration of hepatocyte entry into S-phase and mitosis and earlier expression of genes involved in regulating cell cycle progression (19Ye H. Holterman A. Yoo K.W. Franks R.R. Costa R.H. Mol. Cell. Biol. 1999; 19: 8570-8580Crossref PubMed Scopus (167) Google Scholar, 22Wang X. Hung N.-J. Costa R.H. Hepatology. 2001; 33: 1404-1414Crossref PubMed Scopus (80) Google Scholar). Recent studies indicate that diminished expression of FoxM1B and its cell cycle target genes is associated with reduced proliferation in regenerating hepatocytes of 12-month-old (old-aged) CD-1 mice (18Wang X. Quail E. Hung N.-J. Tan Y., Ye, H. Costa R.H. Proc. Natl. Acad. Sci. U. S. A. 2001; 98: 11468-11473Crossref PubMed Scopus (176) Google Scholar) and in proliferating human fibroblasts of the elderly (23Ly D.H. Lockhart D.J. Lerner R.A. Schultz P.G. Science. 2000; 287: 2486-2492Crossref PubMed Scopus (499) Google Scholar). Maintaining FoxM1B expression in regenerating liver of old-aged TTR-FoxM1B TG mice is sufficient to increase hepatocyte DNA synthesis and mitosis to levels found in young (2-month-old) regenerating mouse liver (18Wang X. Quail E. Hung N.-J. Tan Y., Ye, H. Costa R.H. Proc. Natl. Acad. Sci. U. S. A. 2001; 98: 11468-11473Crossref PubMed Scopus (176) Google Scholar). Furthermore, FoxM1B-mediated stimulation of hepatocyte proliferation was associated with increased expression of numerous cell cycle regulatory genes. These include increased expression of S-phase promoting Cyclin D1 and Cyclin A2 and M-phase promoting Cyclin F, Cyclin B1, Cyclin B2, Cdc25B, and p55CDC (18Wang X. Quail E. Hung N.-J. Tan Y., Ye, H. Costa R.H. Proc. Natl. Acad. Sci. U. S. A. 2001; 98: 11468-11473Crossref PubMed Scopus (176) Google Scholar). Taken together, these published studies supported the hypothesis that maintaining FoxM1B levels will prevent age related decrease in cellular proliferation. In this study, we showed that regenerating livers of old-aged Balb/c mice exhibited diminished FoxM1B expression with significant reductions in hepatocyte proliferation, increased levels of the S-phase inhibitor p27Kip1 (p27) protein, and reduced expression of the M-phase promoting CyclinA2, Cyclin B1, Cyclin B2, Cdc25B, Cdk1, and p55CDC genes. We demonstrated that an adenovirus-mediated increase in FoxM1B expression (AdFoxM1B) significantly elevated hepatocyte proliferation in regenerating liver of old-aged Balb/c mice, which was associated with diminished nuclear expression of the p27 protein and restoring levels of these M-phase progression genes. To create the adenovirus expressing FoxM1B (AdFoxM1B), we subcloned the 2.7-kbEcoRI-HindIII fragment of the human FoxM1B cDNA into the adenovirus shuttle vector pGEMCMV NEW (a gift from J. R. Nevins, Duke University, Durham, NC). This adenovirus shuttle vector and the overlapping adenovirus genome were transfected into 293 cells and homologous recombination was used to generate AdFoxM1B as described previously (24Schwarz J.K. Bassing C.H. Kovesdi I. Datto M.B. Blazing M. George S. Wang X.F. Nevins J.R. Proc. Natl. Acad. Sci. U. S. A. 1995; 92: 483-487Crossref PubMed Scopus (154) Google Scholar). Replication-defective adenovirus, which was deleted in the E1 genes (AdEmpty), was a gift from P. Raychaudhuri (University of Illinois at Chicago, Chicago, IL). Recombinant adenoviruses were used to infect QBO-293 cells (Quantum Biotechnologies, Montreal, Canada), and cell lysates were harvested at 72 h post infection. Adenovirus particles were purified from this cell lysate by CsCl centrifugation and dialyzed to remove the CsCl as described previously (25Tan Y. Costa R.H. Kovesdi I. Reichel R.R. Gene Expr. 2001; 9: 237-248Crossref PubMed Scopus (29) Google Scholar, 26Tan Y. Hughes D.E. Wang X. Costa R.H. Hepatology. 2002; 35: 30-39Crossref PubMed Scopus (24) Google Scholar). Two days prior to the PHx operation, 12-month-old Balb/c mice (purchased from the National Institutes of Aging, Bethesda, MD) were subjected to tail vein injection with 200 μl of either phosphate-buffered saline (PBS; mock-infected) or PBS containing 1 × 1011 purified adenovirus particles (AdFoxM1B or AdEmpty). Two days after infection, all the Balb/c mice were subjected to PHx to induce liver regeneration as described previously (18Wang X. Quail E. Hung N.-J. Tan Y., Ye, H. Costa R.H. Proc. Natl. Acad. Sci. U. S. A. 2001; 98: 11468-11473Crossref PubMed Scopus (176) Google Scholar, 19Ye H. Holterman A. Yoo K.W. Franks R.R. Costa R.H. Mol. Cell. Biol. 1999; 19: 8570-8580Crossref PubMed Scopus (167) Google Scholar). Three mice at each time point were sacrificed using CO2 asphyxiation at the following intervals after PHx: 24, 28, 32, 36, 40, and 44 h. An intraperitoneal injection of a PBS solution containing 10 mg/ml of 5-bromo-2′-deoxyuridine (BrdUrd, Sigma; 50 μg/g body weight) was administered 2 h prior to harvesting the remnant regenerating liver. The regenerating livers were harvested and divided into three portions: one to isolate total RNA (19Ye H. Holterman A. Yoo K.W. Franks R.R. Costa R.H. Mol. Cell. Biol. 1999; 19: 8570-8580Crossref PubMed Scopus (167) Google Scholar), one to isolate total protein extract (27Rausa F.M. Tan Y. Zhou H. Yoo K. Stolz D.B. Watkins S. Franks R.R. Unterman T.G. Costa R.H. Mol. Cell. Biol. 2000; 20: 8264-8282Crossref PubMed Scopus (99) Google Scholar), and one utilized for paraffin embedding (28Rausa F.M., Ye, H. Lim L. Duncan S.A. Costa R.H. Methods. 1998; 16: 29-41Crossref PubMed Scopus (22) Google Scholar). Determination of the number of hepatocytes undergoing DNA synthesis was performed by monoclonal antibody detection of BrdUrd incorporation (Roche) of regenerating liver (5-μm paraffin sections) using the microwave antigen retrieval method described previously (19Ye H. Holterman A. Yoo K.W. Franks R.R. Costa R.H. Mol. Cell. Biol. 1999; 19: 8570-8580Crossref PubMed Scopus (167) Google Scholar). Using three regenerating livers per time point, we counted the number of BrdUrd-positive nuclei per 1000 hepatocytes to calculate the mean number of BrdUrd-positive cells (± S.D.) as described previously (19Ye H. Holterman A. Yoo K.W. Franks R.R. Costa R.H. Mol. Cell. Biol. 1999; 19: 8570-8580Crossref PubMed Scopus (167) Google Scholar). Three regenerating liver sections at 36, 40, and 44 h post-PHx were stained with Hematoxylin and Eosin and examined for mitotic figures (mitosis). Hepatocyte mitosis is expressed as the mean of the number of mitotic figures found per 1000 hepatocytes ± S.D. as described previously (19Ye H. Holterman A. Yoo K.W. Franks R.R. Costa R.H. Mol. Cell. Biol. 1999; 19: 8570-8580Crossref PubMed Scopus (167) Google Scholar). Adenovirus-infected and mock-infected mice undergoing liver regeneration were given free access to food and water. The FoxM1B (HFH-11B) antibody specific to the amino terminus (19Ye H. Holterman A. Yoo K.W. Franks R.R. Costa R.H. Mol. Cell. Biol. 1999; 19: 8570-8580Crossref PubMed Scopus (167) Google Scholar, 21Ye H. Kelly T.F. Samadani U. Lim L. Rubio S. Overdier D.G. Roebuck K.A. Costa R.H. Mol. Cell. Biol. 1997; 17: 1626-1641Crossref PubMed Scopus (315) Google Scholar), p27 antibody (Cell Signaling, Berkeley, CA), or Cdc25B antibody (Santa Cruz Biotechnology, Santa Cruz, CA) was used for immunohistochemical detection of paraffin-embedded 5 μm sections of regenerating liver using methods described previously (19Ye H. Holterman A. Yoo K.W. Franks R.R. Costa R.H. Mol. Cell. Biol. 1999; 19: 8570-8580Crossref PubMed Scopus (167) Google Scholar, 21Ye H. Kelly T.F. Samadani U. Lim L. Rubio S. Overdier D.G. Roebuck K.A. Costa R.H. Mol. Cell. Biol. 1997; 17: 1626-1641Crossref PubMed Scopus (315) Google Scholar). Prior to FoxM1B immunohistochemical staining, regenerating liver sections were subjected to a 15-min proteinase K antigen retrieval step (20 μg/ml Proteinase K, Invitrogen) in PBS at room temperature and then rinsed in PBS. For Western blot analysis, 50 μg of total liver protein (27Rausa F.M. Tan Y. Zhou H. Yoo K. Stolz D.B. Watkins S. Franks R.R. Unterman T.G. Costa R.H. Mol. Cell. Biol. 2000; 20: 8264-8282Crossref PubMed Scopus (99) Google Scholar) was separated on SDS-PAGE and transferred to a Protran membrane (Schleicher & Schuell, Keene, NH). The signal from the primary p27Kip1antibody (Cell Signaling) was amplified by biotin-conjugated anti-rabbit IgG (Bio-Rad, Hercules, CA), and signals were detected with Enhanced Chemiluminescence Plus (ECL-plus, AmershamBiosciences). Cdk2 kinase assays were performed by immunoprecipitation of active Cdk2 enzyme from 200 μg of total liver protein with Cdk2 antibody (Santa Cruz Biotechnology.) and Protein A-Sepharose beads (Amersham Biosciences), and nonspecific proteins were removed by repetitive washes as described in Kiyokawa et al.(29Kiyokawa H. Kineman R.D. Manova-Todorova K.O. Soares V.C. Hoffman E.S. Ono M. Khanam D. Hayday A.C. Frohman L.A. Koff A. Cell. 1996; 85: 721-732Abstract Full Text Full Text PDF PubMed Scopus (1150) Google Scholar). The Cdk2 kinase reaction involved addition of the Rb protein (Santa Cruz Biotechnology) with [γ-32P]ATP to immunoprecipitated Cdk2 protein bound to the Protein A-Sepharose beads. The kinase reaction was incubated for 30 min at 37 °C, and one-half of the Cdk2 kinase reaction was separated by SDS/PAGE and exposed to a phosphorimaging screen. Quantitation of Cdk2-mediated Rb phosphorylation was performed with the Storm 860 PhosphorImager and the ImageQuant program (AmershamBiosciences). Total RNA was prepared from mouse liver at indicated hours post-PHx using RNA-STAT-60 (Tel-Test “B” Inc., Friendswood, TX) and used for RNase protection assays with antisense [α-32P]UTP-labeled probes specific to cell cycle regulatory genes as described previously (18Wang X. Quail E. Hung N.-J. Tan Y., Ye, H. Costa R.H. Proc. Natl. Acad. Sci. U. S. A. 2001; 98: 11468-11473Crossref PubMed Scopus (176) Google Scholar, 22Wang X. Hung N.-J. Costa R.H. Hepatology. 2001; 33: 1404-1414Crossref PubMed Scopus (80) Google Scholar). The 600-nucleotide mouse p27 cDNA (a gift from H. Kiyokawa) was digested withAvaII, and a 200-nucleotide antisense RNA probe was synthesized from this template. RNase protection assay gels were also exposed to phosphorimaging screens for 1 or 2 days and scanned with a Storm 860 PhosphorImager and quantitated with the ImageQuant program. Expression levels were normalized to either cyclophilin or glyceraldehyde-3-phosphate dehydrogenase mRNA levels. Mean mRNA levels ± S.D. were determined from three distinct regenerating livers. Maintaining hepatocyte FoxM1B levels in 12-month-old (old-aged) TTR-FoxM1B TG CD-1 mice stimulated both proliferation of regenerating hepatocytes and expression of cell cycle regulatory genes (18Wang X. Quail E. Hung N.-J. Tan Y., Ye, H. Costa R.H. Proc. Natl. Acad. Sci. U. S. A. 2001; 98: 11468-11473Crossref PubMed Scopus (176) Google Scholar). Because these TG CD-1 mice sustained FoxM1B expression throughout the aging process, we sought to determine whether an acute adenovirus-mediated delivery of hepatic FoxM1B protein (AdFoxM1B) to old-aged Balb/c mice would stimulate proliferation of regenerating hepatocytes. To avoid the initial acute phase response to viral infection, which subsides within 36 h following adenovirus infection (30Benihoud K. Salone B. Esselin S. Opolon P. Poli V., Di Giovine M. Perricaudet M. Saggio I. J. Gene Med. 2000; 2: 194-203Crossref PubMed Scopus (30) Google Scholar), we chose to perform the PHx two days after adenovirus infection. 12-month-old (old-aged) Balb/c mice were infected with either replication-defective control adenovirus (AdEmpty) or AdFoxM1B by tail vein injection and then subjected to PHx operations two days later. The remnant regenerating livers were harvested at different intervals between 24 and 44 h following surgery for analysis of hepatocyte proliferation and expression of cell cycle genes. Hepatocyte DNA synthesis was monitored by immunohistochemical staining of BrdUrd incorporation into DNA as described previously (18Wang X. Quail E. Hung N.-J. Tan Y., Ye, H. Costa R.H. Proc. Natl. Acad. Sci. U. S. A. 2001; 98: 11468-11473Crossref PubMed Scopus (176) Google Scholar). RNase protection assays with regenerating liver RNA demonstrated that only AdFoxM1B-infected old-aged mice exhibited increased hepatic FoxM1B mRNA that resembled levels observed with young (2-month-old) mouse liver at 40 h after PHx (Fig.1 A). AdFoxM1B infection of old-aged mice increased the peak of regenerating hepatocyte DNA replication and mitosis (Fig. 1, B and C) compared with mock-infected (MI) or AdEmpty-infected adenovirus controls (Fig. 1, B and C). Furthermore, AdFoxM1B-infected old-aged mice displayed an 8-h acceleration in the peak of hepatocyte DNA replication occurring at 32 h post-PHx (Fig. 1 B, red line) compared with a DNA replication peak at 40 h after PHx found with young regenerating liver (Fig. 1 B, black line). This accelerated DNA replication peak was associated with earlier hepatocyte FoxM1B nuclear staining at 24 h post-PHx (Fig.2 A) compared with 36–40 h post-PHx for young regenerating mouse liver (Fig. 2 B). In contrast, no FoxM1B nuclear staining was found in either MI- or AdEmpty-infected old-aged mouse controls (Fig. 2, C andD). These results suggest that earlier FoxM1B nuclear staining is associated with accelerated hepatocyte DNA replication in regenerating liver of AdFoxM1B-infected old-aged mice.Figure 2Regenerating liver of AdFoxM1B-infected old-aged mice displayed earlier nuclear FoxM1B staining.Regenerating liver sections were subjected to immunohistochemical staining with an amino-terminal FoxM1B antibody. This includes regenerating liver from either AdFoxM1B-infected 12-month-old (12 M) (A), mock-infected (MI) 2-month-old (B), MI 12-month-old (C), or AdEmpty-infected 12-month-old (D) Balb/c mice. Shown are regenerating liver sections isolated at 24, 32, or 40 h following PHx.Arrows indicate representative FoxM1B nuclear staining. Magnification for panels A, C, and Dare 400× and panel B is 630×.View Large Image Figure ViewerDownload (PPT) The p27Kip1 (p27) protein associates with Cdk2 and inhibits kinase activity of the Cyclin E-Cdk2 and Cyclin A2-Cdk2 complexes (11Sherr C.J. Roberts J.M. Genes Dev. 1999; 13: 1501-1512Crossref PubMed Scopus (5158) Google Scholar). S-phase progression requires Cyclin-Cdk2 protein phosphorylation of the RB protein, which causes dissociation of RB and activates the E2F transcription factor (6Harbour J.W. Dean D.C. Genes Dev. 2000; 14: 2393-2409Crossref PubMed Scopus (963) Google Scholar). Because the related FoxO1 and FoxO3 (previously called FKHR) transcription factors are known to stimulate expression of p27 (31Medema R.H. Kops G.J. Bos J.L. Burgering B.M. Nature. 2000; 404: 782-787Crossref PubMed Scopus (1231) Google Scholar), we next examined whether AdFoxM1B infection could influence expression of the p27 protein. We subjected regenerating liver protein extracts to Western blot analysis using a commercially available p27 antibody (Fig.3). These data demonstrated that either MI- or AdEmpty-infected old-aged mice displayed an elevated expression of p27 protein (Fig. 3, B and D), whereas reduced hepatic p27 levels were found in both young mice and AdFoxM1B-infected old-aged mice following PHx (Fig. 3, A and C). Interestingly, we found that AdFoxM1B infection did not alter p27 mRNA levels (Fig. 3, E and F), suggesting that FoxM1B is regulating p27 protein levels rather than controlling its promoter activity. Furthermore, immunohistochemical staining of regenerating liver sections demonstrated that MI- and AdEmpty-infected old-aged mice displayed abundant p27 nuclear staining prior to S-phase (Fig. 4, A, B,D, and E). In contrast, AdFoxM1B-infected old-aged mice displayed only perinuclear hepatocyte staining of p27 protein following PHx (Fig. 4, G–I), a finding consistent with stimulating S-phase progression (11Sherr C.J. Roberts J.M. Genes Dev. 1999; 13: 1501-1512Crossref PubMed Scopus (5158) Google Scholar). Furthermore, although MI- and AdEmpty-infected old-aged mice displayed perinuclear staining of p27 protein at 40 h post-PHx (Fig. 4, C andF), this delayed change in p27 cellular localization was unable to facilitate S-phase progression. Collectively, these liver regeneration studies suggest that FoxM1B mediates S-phase progression in old-aged mice by diminishing nuclear expression of the p27 protein.Figure 4AdFoxM1B-infected old-aged mice display perinuclear p27 staining in regenerating hepatocytes. Regenerating liver sections were subjected to immunohistochemical staining with a p27 antibody. This includes regenerating liver from either MI 12-month-old (A–C), AdEmpty-infected 12-month-old (D–F), or AdFoxM1B-infected 12-month-old (G–I) Balb/c mice. Shown are regenerating liver sections isolated at 24 or 28, 36, and 40 h following PHx. Note that the DNA replication peak occurs at 32 h post-PHx in AdFoxM1B-infected old-aged mice, whereas it occurs at 40 h post-PHx in either MI- or AdEmpty-infected old-aged mice.Arrows indicate representative p27 staining, which is either nuclear in MI- and AdEmpty-infected old-aged mice or perinuclear in AdFoxM1B-infected mice.View Large Image Figure ViewerDownload (PPT) To examine whether diminished p27 protein levels were associated with increased Cdk2 kinase activity, regenerating liver protein extracts prepared from either AdEmpty- or AdFoxM1B-infected old-aged mice were immunoprecipitated with Cdk2 antibody and then used to phosphorylate recombinant RB protein. These experiments demonstrated that diminished S-phase levels of p27 protein were associated with elevated Cdk2 kinase activity in AdFoxM1B-infected regenerating liver (Fig.5 B) compared with AdEmpty-infected controls (Fig. 5 A). These results suggest that AdFoxM1B infection of old-aged mice diminished nuclear expression of p27 protein in regenerating hepatocytes, which allowed complex formation of the active Cyclin E-Cdk2 and Cyclin A-Cdk2 complexes required for S-phase progression. RNase protection assays were performed in triplicate with cell cycle regulatory gene probes, and regenerating liver RNA was isolated between 24 and 44 h following PHx (Fig.6, A–D). In 2-month-old (young) regenerating mouse liver, increased S-phase levels of Cyclin A2, Cyclin B1, and Cyclin B2 mRNA occurred at 40 h post-PHx (Fig. 6, E–G). We found that regenerating liver of AdFoxM1B-infected old-aged mice displayed increased S-phase levels of Cyclin A2, Cyclin B1, and Cyclin B2 mRNA at 32 h post-PHx (Fig. 6, E–G) compared with either MI- or AdEmpty-infected old-aged mouse controls (Fig. 6, E–G). Furthermore, expression levels of these genes diminished during mitosis, which occurred at 36 h following PHx (Fig. 6, E–G). In addition, expression of mitosis promoting Cdc25B phosphatase, Cdk1, and p55CDC mRNA initiated prior to S-phase between 32 and 36 h following PHx in young regenerating liver (Fig.7 B, arrow). Likewise, AdFoxM1B-infected regenerating liver initiated expression of Cdc25B, p55CDC, and Cdk1 mRNA prior to S-phase at 24 h after PHx (Fig. 7 D, arrow), whereas these mRNA levels were diminished in regenerating liver of either MI- or AdEmpty-infected old-aged mouse controls (Fig. 7, A andC). Immunohistochemical staining of regenerating liver sections revealed a biphasic nuclear staining of Cdc25B phosphatase protein prior to and following S-phase (Fig.8, D–F), a finding consistent with the ability of Cdc25B to activate Cdk1 as required for M-phase progression (8Nilsson I. Hoffmann I. Prog. Cell Cycle Res. 2000; 4: 107-114Crossref PubMed Scopus (384) Google Scholar, 9Sebastian B. Kakizuka A. Hunter T. Proc. Natl. Acad. Sci. U. S. A. 1993; 90: 3521-3524Crossref PubMed Scopus (213) Google Scholar, 10Trembley J.H. Ebbert J.O. Kren B.T. Steer C.J. Cell Growth & Differ. 1996; 7: 903-916PubMed Google Scholar). In contrast, low levels of either nuclear or perinuclear hepatocyte staining of Cdc25B protein were found at 28 h post-PHx in AdEmpty-infected old-aged liver (Fig.8 A). Furthermore, only perinuclear hepatocyte staining of the Cdc25B protein was found at later time points following PHx (Fig.8, A–C, and data not shown). These results suggest that FoxM1B-stimulated proliferation of regenerating hepatocytes is associated with increased expression of S-phase promoting Cyclin A genes and M-phase promoting Cyclin B1, Cyclin B2, Cdc25B, and p55CDC genes.Figure 7Increased Cdc25B, p55CDC, and Cdk1 mRNA levels during liver regeneration of AdFoxM1B-infected old-aged mice. Total RNA was isolated from regenerating liver and analyzed for RNA expression by RNase protection assays (in triplicate) with Cdc25B, p55CDC, and Cdk1 probes. This includes regenerating liver from MI 12-month-old (A), MI 2-month-old (B), AdEmpty-infected 12-month-old (C), or AdFoxM1B-infected 12-month old (D) Balb/c mice. Note that representative duplicate samples are shown. Expression levels were normalized to cyclophilin internal control. The relative fold induction of expression is indicated below the panels for Cdk1. Expression levels of the MI 2-month-old regenerating mouse liver at 24 h post-PHx were set at 1.0. Note that Cdc25B and p55CDC levels were undetectable in regenerating livers from either MI- or AdEmpty-infected 12-month-old mice. Arrows indicate the onset of expression of p55CDC, Cdc25B, and Cdk-1 during liver regeneration.View Large Image Figure ViewerDownload (PPT)Figure 8Increased hepatocyte nuclear staining of Cdc25B protein prior to S-phase in liver regeneration of AdFoxM1B-infected old-aged mice. Regenerating liver sections were subjected to immunohistochemical staining with a Cdc25B antibody. This includes regenerating liver from either AdEmpty-infected 12-month-old (A–C) or AdFoxM1B-infected 12-month-old (D–F) Balb/c mice. Shown are regenerating liver sections isolated at 28, 32, or 40 h following PHx.Arrows indicate representative Cdc25B nuclear staining.View Large Image Figure ViewerDownload (PPT) Previous liver regeneration studies demonstrated that maintaining hepatic expression of FoxM1B in 12-month-old (old-aged) TTR-FoxM1B TG mice is sufficient to increase regenerating hepatocyte proliferation and expression of numerous cell cycle regulatory genes (18Wang X. Quail E. Hung N.-J. Tan Y., Ye, H. Costa R.H. Proc. Natl. Acad. Sci. U. S. A. 2001; 98: 11468-11473Crossref PubMed Scopus (176) Google Scholar). In this study we demonstrate that AdFoxM1B infection of old-aged mice prior to PHx elevated FoxM1B expression levels and increased hepatocyte DNA replication and mitosis compared with either MI- or AdEmpty-infected controls. This stimulation in hepatocyte S-phase progression was associated with diminished expression of the Cdk inhibitor p27Kip1 (p27) protein and sustained perinuclear staining of the p27 protein. In contrast, old-aged mice infected with control virus displayed high hepatocyte levels of p27 protein, which had been localized to the nucleus prior to S-phase. Interestingly, we found that AdFoxM1B infection did not alter p27 mRNA levels, a new result demonstrating that FoxM1B participates in regulating nuclear expression of the p27 protein rather than controlling its promoter expression. Furthermore, we found increased S-phase levels of Cyclin A2 mRNA and Cdk2 kinase activity, both of which are consistent with enhanced hepatocyte DNA replication (7Zachariae W. Nasmyth K. Genes Dev. 1999; 13: 2039-2058Crossref PubMed Scopus (576) Google Scholar, 11Sherr C.J. Roberts J.M. Genes Dev. 1999; 13: 1501-1512Crossref PubMed Scopus (5158) Google Scholar). Moreover, AdFoxM1B infection of old-aged mice caused elevated expression of Cyclin B1, Cyclin B2, Cdc25B, Cdk1, and p55CDC mRNA in regenerating livers, all of which are required for mitosis (7Zachariae W. Nasmyth K. Genes Dev. 1999; 13: 2039-2058Crossref PubMed Scopus (576) Google Scholar). Finally, we showed that increased FoxM1B levels in regenerating old-aged hepatocytes enhanced nuclear staining of Cdc25B protein, a phosphatase, which stimulated Cdk1 activity required for entry into mitosis (8Nilsson I. Hoffmann I. Prog. Cell Cycle Res. 2000; 4: 107-114Crossref PubMed Scopus (384) Google Scholar, 9Sebastian B. Kakizuka A. Hunter T. Proc. Natl. Acad. Sci. U. S. A. 1993; 90: 3521-3524Crossref PubMed Scopus (213) Google Scholar, 10Trembley J.H. Ebbert J.O. Kren B.T. Steer C.J. Cell Growth & Differ. 1996; 7: 903-916PubMed Google Scholar). These studies indicated that an acute delivery of FoxM1B gene in old-aged mice was sufficient to re-establish regenerating hepatocyte proliferation, suggesting that FoxM1B is a candidate for therapeutic intervention to ameliorate diminished proliferation observed in the elderly. Activation of Cdk activity through the formation of CyclinD1-Cdk 4/6 and Cyclin E-Cdk2 complexes is required to phosphorylate the RB protein and mediate release of the E2F transcription factor to activate expression of its proliferation-specific target genes (6Harbour J.W. Dean D.C. Genes Dev. 2000; 14: 2393-2409Crossref PubMed Scopus (963) Google Scholar). Unlike liver regeneration studies with old-aged wild type and TG CD-1 mice (18Wang X. Quail E. Hung N.-J. Tan Y., Ye, H. Costa R.H. Proc. Natl. Acad. Sci. U. S. A. 2001; 98: 11468-11473Crossref PubMed Scopus (176) Google Scholar), we did not find significant differences in Cyclin D1 and Cyclin E expression between old-aged and young Balb/c mice. However, AdFoxM1B infection was able to stimulate S-phase levels of Cyclin A2 mRNA and diminish nuclear expression of the p27 protein, which was elevated in old-aged mouse liver (Figs. 3 and 4). A transient reduction in p27 protein levels is required for activation of Cyclin E-Cdk2 and Cyclin A2-Cdk2 kinase complex, which is essential to mediate progression into S-phase through maintenance of RB phosphorylation and E2F transcriptional activity (6Harbour J.W. Dean D.C. Genes Dev. 2000; 14: 2393-2409Crossref PubMed Scopus (963) Google Scholar). Consistent with this finding, AdFoxM1B-infected regenerating liver displayed increased S-phase Cdk2 activity using the RB protein substrate compared with AdEmpty-infected old-aged mouse controls. Moreover, the active Cyclin A2-Cdk2 kinase complex is required to phosphorylate the cdh1 subunit of the ubiquitin-ligase APC, which prevents APC-mediated degradation of Cyclin B at the end of S-phase and thus allows Cyclin B accumulation to promote entry into mitosis (6Harbour J.W. Dean D.C. Genes Dev. 2000; 14: 2393-2409Crossref PubMed Scopus (963) Google Scholar). More recent studies demonstrate that in proliferating mammalian cells the phosphatidylinositol 3′-kinase (PI3K)/Akt signal transduction pathway is essential for G1 to S-phase progression because it prevents transcriptional activity of the FoxO1 and FoxO3 (previously called FKHR) proteins, which stimulate transcription of the Cdk inhibitor p27Kip1 gene (31Medema R.H. Kops G.J. Bos J.L. Burgering B.M. Nature. 2000; 404: 782-787Crossref PubMed Scopus (1231) Google Scholar). Activation of the PI3K signaling pathway stimulates the Akt kinase, which phosphorylates the C-terminus of the FoxO1 (Fkhr; Daf-16) protein and mediates its nuclear export into the cytoplasm, thus preventing FoxO1 transcriptional activation of target genes (32Biggs III, W.H. Meisenhelder J. Hunter T. Cavenee W.K. Arden K.C. Proc. Natl. Acad. Sci. U. S. A. 1999; 96: 7421-7426Crossref PubMed Scopus (945) Google Scholar, 33Brunet A. Bonni A. Zigmond M.J. Lin M.Z. Juo P., Hu, L.S. Anderson M.J. Arden K.C. Blenis J. Greenberg M.E. Cell. 1999; 96: 857-868Abstract Full Text Full Text PDF PubMed Scopus (5454) Google Scholar, 34Guo S. Rena G. Cichy S., He, X. Cohen P. Unterman T. J. Biol. Chem. 1999; 274: 17184-17192Abstract Full Text Full Text PDF PubMed Scopus (469) Google Scholar). Unlike the FoxO protein, whose transcriptional inhibition is required for cell cycle progression, elevated FoxM1B expression mediates S-phase progression by diminishing nuclear expression of the p27 protein. Interestingly, increased FoxM1B expression did not alter p27 mRNA levels, indicating that FoxM1B regulates p27 protein levels rather than influencing its promoter activity. During G1 phase of the cell cycle, the p27 protein is negatively regulated by phosphorylation, which mediates association with the Jab1 protein, which mediates p27 nuclear export using a CRM1-dependent mechanism and its subsequent proteosome-mediated degradation (35Hara T. Kamura T. Nakayama K. Oshikawa K. Hatakeyama S. J. Biol. Chem. 2001; 276: 48937-48943Abstract Full Text Full Text PDF PubMed Scopus (197) Google Scholar, 36Rodier G. Montagnoli A., Di Marcotullio L. Coulombe P. Draetta G.F. Pagano M. Meloche S. EMBO J. 2001; 20: 6672-6682Crossref PubMed Scopus (252) Google Scholar, 37Tomoda K. Kubota Y. Arata Y. Mori S. Maeda M. Tanaka T. Yoshida M. Yoneda-Kato N. Kato J.Y. J. Biol. Chem. 2002; 277: 2302-2310Abstract Full Text Full Text PDF PubMed Scopus (253) Google Scholar, 38Tomoda K. Kubota Y. Kato J. Nature. 1999; 398: 160-165Crossref PubMed Scopus (551) Google Scholar). Interestingly, we find that restoring FoxM1B expression causes perinuclear hepatocyte staining of the p27 protein in old-aged regenerating liver. It is therefore tempting to speculate that FoxM1B causes diminished p27 protein levels through CRM-1-mediated p27 nuclear export and degradation. Previous co-transfection studies in tissue culture cells demonstrated that FoxM1B over-expression could activate transcription of the Cyclin D1 and Cyclin B1 promoters (18Wang X. Quail E. Hung N.-J. Tan Y., Ye, H. Costa R.H. Proc. Natl. Acad. Sci. U. S. A. 2001; 98: 11468-11473Crossref PubMed Scopus (176) Google Scholar). Our present study shows, for the first time, that FoxM1B is capable of directly or indirectly regulating nuclear expression of the p27 protein, which potentiates formation of the Cyclin-Cdk2 complexes required to stimulate hepatocyte S-phase progression. Consistent with our previous liver regeneration studies in old-aged TG mice, increased FoxM1B levels stimulated expression of M-phase promoting Cyclin B1, Cyclin B2, and Cdk1 genes. Cyclin B proteins associate with Cdk1 to mediate cell cycle progression from the G2 phase into mitosis (7Zachariae W. Nasmyth K. Genes Dev. 1999; 13: 2039-2058Crossref PubMed Scopus (576) Google Scholar). Furthermore, AdFoxM1B infection of regenerating old-aged liver increased expression of Cdc25B and p55CDC genes, which was undetectable in either MI- or AdEmpty-infected controls. Elevated FoxM1B levels were associated with increased Cdc25B nuclear staining of regenerating hepatocytes, a new finding consistent with stimulating M-phase progression. The Cdc25B phosphatase promotes M-phase progression by dephosphorylation of the Cdk1 Tyr-15 and Thr-14 residues, thereby activating the mitotic Cdk1/Cyclin B kinase (8Nilsson I. Hoffmann I. Prog. Cell Cycle Res. 2000; 4: 107-114Crossref PubMed Scopus (384) Google Scholar, 9Sebastian B. Kakizuka A. Hunter T. Proc. Natl. Acad. Sci. U. S. A. 1993; 90: 3521-3524Crossref PubMed Scopus (213) Google Scholar, 10Trembley J.H. Ebbert J.O. Kren B.T. Steer C.J. Cell Growth & Differ. 1996; 7: 903-916PubMed Google Scholar). Microinjection of the Cdc25B protein, but not the Cdc25C protein, can drive replicating cells into mitosis, suggesting that Cdc25B regulates entry into mitosis (39Karlsson C. Katich S. Hagting A. Hoffmann I. Pines J. J. Cell Biol. 1999; 146: 573-584Crossref PubMed Scopus (147) Google Scholar). The fact that increased FoxM1B levels re-established appropriate increased expression of Cdc25B is consistent with the ability of FoxM1B to restore mitosis in regenerating liver of old-aged mice. Finally, FoxM1B-mediated increase in p55CDC levels is required to exit mitosis through its regulation of the ubiquitin-ligase APC degradation of the Cyclin proteins (7Zachariae W. Nasmyth K. Genes Dev. 1999; 13: 2039-2058Crossref PubMed Scopus (576) Google Scholar). In summary, reduced expression of the FoxM1B transcription factor contributes to the decline in regenerating hepatocyte proliferation observed in old-aged mice. We show that adenovirus delivery of the FoxM1B gene was sufficient to restore FoxM1B expression and stimulated hepatocyte proliferation in regenerating liver of old-aged mice. Increased FoxM1B levels diminished nuclear p27 levels and stimulated expression of the necessary cell cycle regulatory genes. We thank H. Kiyokawa for assistance with the Cdk2 kinase assays and Nai-Jung Hung for technical assistance. We also thank P. Raychaudhuri, V. Kalinichenko, Y. Zhou, M. Major, and F. Rausa for critically reviewing the manuscript.