SEARCH

SEARCH BY CITATION

Keywords:

  • winged helix DNA binding domain;
  • Forkhead transcription factor;
  • heart development;
  • Trident;
  • cardiomyocyte;
  • p21;
  • knockout mice;
  • NFATc3

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES

The Forkhead Box m1 (Foxm1) transcription factor is expressed in cardiomyocytes and cardiac endothelial cells during heart development. In this study, we used a novel Foxm1 −/− mouse line to demonstrate that Foxm1-deletion causes ventricular hypoplasia and diminished DNA replication and mitosis in developing cardiomyocytes. Proliferation defects in Foxm1 −/− hearts were associated with a reduced expression of Cdk1-activator Cdc25B phosphatase and NFATc3 transcription factor, and with abnormal nuclear accumulation of the Cdk-inhibitor p21Cip1 protein. Depletion of Foxm1 levels by siRNA caused altered expression of these genes in cultured HL-1 cardiomyocytes. Endothelial-specific deletion of the Foxm1 fl/fl allele in Tie2-Cre Foxm1 fl/fl embryos did not influence heart development and cardiomyocyte proliferation. Foxm1 protein binds to the −9,259/−9,288-bp region of the endogenous mouse NFATc3 promoter, indicating that Foxm1 is a transcriptional activator of the NFATc3 gene. Foxm1 regulates expression of genes essential for the proliferation of cardiomyocytes during heart development. Developmental Dynamics 236:1000–1013, 2007. © 2007 Wiley-Liss, Inc.


INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES

Development of mouse hearts initiates at 7.0 days post coitum (dpc) when the cardiac lineage becomes specified from cells of the primitive streak (Zaffran and Frasch, 2002). The cardiac mesoderm migrates anterolaterally and fuses at the ventral aspect of the embryo to form a linear heart tube, which subsequently undergoes complex looping and septation resulting in the formation of the four-chambered heart (Zaffran and Frasch, 2002; Eisenberg and Markwald, 2004). The transcriptional network, which controls the development of mammalian hearts, has been extensively studied. Disruption of this transcriptional network results in a variety of human diseases including congenital heart disease, dilated cardiomyopathy, and cardiac hypertrophy (Frey and Olson, 2003; Olson, 2004; Clark et al., 2006). Central to the regulation of these developmental processes are the cardiac transcription factors GATA4 (Watt et al., 2004), NKX2.5 (Cripps and Olson, 2002), Hand (McFadden et al., 2005), nuclear factors of activated T-cell (NFAT) c3 and c4 (Molkentin, 2000; Graef et al., 2001; Bushdid et al., 2003), Foxp4 (Li et al., 2004) and FOG2 (Svensson et al., 2000; Lin et al., 2004), and T-box transcription factors TBX1 and TBX5 (Plageman and Yutzey, 2005), all of which have been shown to regulate cardiomyocyte proliferation, cardiac-specific gene expression, and proper patterning of the developing heart.

The Forkhead Box (Fox) proteins are an extensive family of transcription factors, which share homology in the Winged Helix/Forkhead DNA binding domain (Clark et al., 1993; Clevidence et al., 1993; Kaestner et al., 1993). Expression of the Foxm1 transcription factor (previously known as HFH-11B, Trident, Win, or MPP2) is induced during cellular proliferation in a variety of different cell types and extinguished in terminally differentiated cells (Korver et al., 1997; Ye et al., 1997). Liver regeneration studies demonstrated that mice with postnatal hepatocyte-specific deletion of the Foxm1 fl/fl (LoxP targeted) allele exhibit a significant reduction in hepatocyte DNA replication and mitosis, which is associated with altered expression of proteins that limit Cdk1 and Cdk2 activity required for normal cell cycle progression (Wang et al., 2002). Recently, we demonstrated that Foxm1 is required for aberrant proliferation of tumor cells during progression of liver, lung, and prostate cancers (Kalinichenko et al., 2004; Kalin et al., 2006; Kim et al., 2006).

Previous studies described the generation of Foxm1−/−neo mouse line that contained a targeted insertion of PGK-neomycin cassette into the third exon of Foxm1 gene without deleting any of the coding sequences (Korver et al., 1998). Although the Foxm1−/−neo mice died postnatally, displaying accumulation of polyploid cardiomyocytes and hepatoblasts (Korver et al., 1998), the authors did not provide any molecular mechanisms of this cardiovascular phenotype and, therefore, the precise role of Foxm1 during heart development remains uncharacterized. We generated a distinct Foxm1 −/− mouse line, which contains a targeted deletion of exons 4 to 7 encoding the Foxm1 winged helix DNA binding and the C-terminal transcriptional activation domains (Krupczak-Hollis et al., 2004). Most of the Foxm1 −/− embryos died in utero between 13.5 and 16.5 dpc due to severe defects in development of the embryonic liver, lung, and heart (Krupczak-Hollis et al., 2004; Kim et al., 2005a). Foxm1 −/− livers displayed abnormal accumulation of polyploid hepatoblasts resulting from diminished DNA replication and a failure to enter mitosis (Krupczak-Hollis et al., 2004). Mouse embryonic fibroblasts (MEF) from our Foxm1 −/− embryos were unable to grow in culture due to a block in mitotic progression and undergo cellular senescence at passage three with high expression levels of senescence-associated β-galactosidase and cell cycle inhibitors p21Cip1, p27Kip1, p16Ink4A, and p19ARF proteins (Wang et al., 2005). This mitotic block was also associated with diminished protein levels of the Polo-like kinase 1 and Aurora B kinase (Krupczak-Hollis et al., 2004; Kim et al., 2005a; Wang et al., 2005), both of which phosphorylate regulatory proteins essential for orchestrating mitosis and cytokinesis (Glover et al., 1998; Adams et al., 2001). Foxm1 is required for differentiation of hepatoblast precursor cells toward billiary epithelial cell lineage because Foxm1 −/− livers fail to develop intrahepatic bile ducts (Krupczak-Hollis et al., 2004). Likewise, Foxm1 −/− embryos exhibit defects in differentiation of pulmonary mesenchyme into mature capillary endothelial cells during the canalicular stage of lung development (Kim et al., 2005a).

In this report, we used our Foxm1 −/− mouse line to investigate the role of Foxm1 in cardiac morphogenesis. We demonstrate that Foxm1 deletion does not affect looping heart morphogenesis or cardiomyocyte differentiation, but reduces both DNA replication and mitosis in developing cardiomyocytes. Proliferation defects in embryonic Foxm1 −/− hearts were associated with reduced expression of Cdk1-activator Cdc25B phosphatase and NFATc3 transcription factor, and with abnormal nuclear accumulation of the Cdk-inhibitor p21Cip1 protein. Likewise, HL-1 cardiomyocytes depleted in Foxm1 levels by siRNA transfection exhibited altered expression of these genes, indicating a direct influence of Foxm1 on cardiomyocyte proliferation. We also showed that Foxm1 binds to the −9,259/−9,288-bp region of endogenous mouse NFATc3 promoter in the developing heart, suggesting that the mouse NFATc3 gene is a direct transcriptional target of Foxm1 during heart development. In contrast to other embryonic tissues, Foxm1 −/− hearts displayed normal expression of many known Foxm1-target genes including Polo-like kinase 1, Aurora B kinase, Skp2, Cks1, and Laminin α4. We also demonstrated that endothelial-specific deletion of the Foxm1 fl/fl allele does not affect heart development or cardiomyocyte proliferation, suggesting that Foxm1 deficiency in endothelial cells does not contribute to proliferation defects in embryonic Foxm1 −/− hearts. These results suggest that the Foxm1 transcription factor is essential for mouse embryonic heart development through regulation of distinct cellular pathways essential for proliferation of cardiomyocytes.

RESULTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES

Foxm1 −/− Embryonic Hearts Exhibited Ventricular Hypoplasia

Paraffin sections from mouse embryos that contained a targeted deletion of exons 4 to 7 of the Foxm1 gene (Foxm1 −/−; Krupczak-Hollis et al., 2004) were stained with hematoxylin and eosin (H&E) and compared with embryonic hearts isolated from aged-matched wild type (WT) embryos to examine for gross morphological defects in heart development. Although no obvious heart abnormalities were observed in Foxm1 −/− 9.5 dpc embryos (Fig. 1A and E), Foxm1 −/− 13.5–18.5-dpc hearts exhibited thinning of ventricular myocardium (Fig. 1B–D, F–H), which is consistent with development of ventricular hypoplasia. Morphometric analysis of embryonic hearts demonstrated that Foxm1 −/− 13.5-dpc hearts displayed a 43% reduction in the thickness of left ventricle compared to WT hearts, whereas a 57% reduction in the thickness of left ventricle was found in Foxm1 −/− 15.5-dpc hearts (Fig. 1R). Although no significant differences in heart size were observed in Foxm1 −/− 13.5-dpc embryos, Foxm1 −/− 15.5-dpc hearts exhibited a 25% reduction in the heart size when compared to WT embryos (Fig. 1Q). Interestingly, Foxm1 −/− 13.5-dpc hearts displayed a significant 1.3-fold increase in thickness of heart valve leaflets compared to WT littermates (Fig. 1J, N; and data not shown).

thumbnail image

Figure 1. Foxm1 −/− embryonic hearts displayed ventricular hypoplasia. A–K, M–O: Gross morphological defects in Foxm1 −/− hearts. Embryonic hearts from WT and Foxm1 −/− 9.5–18.5 dpc embryos were fixed, paraffin-embedded, sectioned, and then stained with hematoxylin and eosin (H&E). Foxm1 −/− embryos displayed dilated ventricles (F,G), thinning of myocardial wall (G,H, arrowheads), and abnormal accumulation of enlarged cardiomyocytes in myocardium (arrows in M and O) and trabecular region (insert in H) compared to WT littermates (A–D and I–K). Foxm1 −/− hearts displayed thickening of atrio-ventricular (A/V) valves (mitral valve leaflets in J and N). L,P: DAPI staining shows an increase in size of cardiomyocyte nuclei (white arrows) in Foxm1 −/− 18.5-dpc hearts. Q–S:Foxm1 −/− 15.5-dpc embryos display reduction in the heart size (Q) and increased size of cardiomyocyte nuclei (S). Diminished thickness of left ventricle is observed in Foxm1 −/− 13.5- and 15.5-dpc hearts (R). Measurements for WT 13.5 dpc heart were taken as 100%. Means ± S.D. were determined using three different WT or Foxm1 −/− embryos. *P < 0.05,**P < 0.01 At, atrium; Ve, ventricle; LA, left atrium; RA, right atrium; LV, left ventricle; RV, right ventricle; A/V, atrio-ventricular valve. Magnifications: ×100 (A–H), ×200 (J,N)m and ×400 (I, K–M, O,P).

Download figure to PowerPoint

Although enlarged cardiomyocytes were found in the Foxm1 −/− atrium and ventricle as early as 9.5 dpc (Fig. 1I and M; and data not shown), severe accumulation of polyploid cardiomyocytes was observed in Foxm1 −/− 13.5–18.5-dpc heart tissues, including ventricular myocardium, interventricular septum (IVS), atrium, and heart trabeculae (Fig. 1D,H, K,L,O,P). No significant differences were found in either length or thickness of heart trabeculae by morphometric analysis of Foxm1 −/− and WT 13.5-dpc hearts (data not shown). These results suggest that Foxm1 is dispensable for trabeculogenesis in the heart. Despite a reduced number of cardiomyocytes in Foxm1 −/− hearts (Fig. 1O,P), they still expressed desmin (Fig. 2E–F), laminin α4 extra-cellular matrix protein (Lama4; Fig. 2I,J), and GATA-4 transcription factor (Fig. 2G,H), all of which are essential for maintaining a differentiated phenotype in cardiomyocyte lineage (Watt et al., 2004; Wang et al., 2006).

thumbnail image

Figure 2. Foxm1 −/− cardiomyocytes display normal expression of GATA4, desmin, and laminin α4. A–D: The Foxm1 expression pattern in the developing heart. Heart paraffin sections from WT 9.5–18.5-dpc embryos were stained with antibodies specific to Foxm1. Foxm1 immunostaining was visualized using biotinylated secondary antibody, avidin-alkaline phosphatase (AP) complex, and BCIP/NBT substrate (dark purple). Slides were counterstained with nuclear fast red (red). Foxm1 is expressed in proliferative cardiomyocytes (arrowheads) and cardiac endothelial cells (arrows). E–J: Heart paraffin sections from WT and Foxm1 −/− 13.5–15.5-dpc embryos were stained with antibodies against desmin (E,F), GATA4 (G,H), or laminin α4 (Lama4; I,J). GATA4 staining was visualized using biotinylated secondary antibody, avidin-HRP, and DAB substrate (brown), and then counterstained with hematoxylin (blue in G,H). Desmin (E,F) and Lama4 (I,J) imunostainings were visualized using biotinylated secondary antibody, avidin-alkaline phosphatase complex, and BCIP/NBT substrate (purple), and then counterstained with nuclear fast red (red). Foxm1 −/− and WT hearts exhibited similar GATA4, desmin, and Lama4 staining. At, atrium; Ve, ventricle; LA, left atrium; LV, left ventricle. Magnifications: ×200 (A,E,F, I,J), ×100 (B), and ×400 (C,D,G,H, and all inserts).

Download figure to PowerPoint

To determine Foxm1 expression pattern in embryonic hearts, paraffin sections from WT hearts were stained with antibodies specific to Foxm1 protein. Foxm1 was expressed in highly proliferative cells of compact layer and trabecular cardiomyocytes of 9.5-dpc WT hearts (Fig. 2A). Abundant Foxm1 nuclear staining was also detected in cardiomyocytes and endothelial cells of 15.5-dpc WT embryos, whereas Foxm1 staining was significantly decreased in 18.5-dpc WT hearts (Fig. 2B–D), correlating with reduced proliferation rates. Foxm1 protein was not detected in adult WT hearts (data not shown).

Foxm1 −/− Hearts Exhibited Diminished DNA Replication and Mitosis

Foxm1 −/− 15.5-dpc embryos displayed significantly reduced numbers of cardiomyocytes (Fig. 3G), suggesting that Foxm1 stimulates cardiomyocyte proliferation during heart development. In order to determine the mechanism(s) of the proliferation defect in Foxm1 −/− cardiomyocytes, we measured DNA replication rates in Foxm1 −/− and WT 15.5-dpc hearts by injecting 5-bromo-2′-deoxyuridine (BrdU) into pregnant female mice 2 hr prior to harvesting the embryos. Immunohistochemistry with BrdU antibody demonstrated that Foxm1 −/− 15.5-dpc hearts exhibited a significant reduction in the number of BrdU-positive cardiomyocytes compared to age-matched WT embryos (Fig. 3A,B,H), whereas DNA replication rates were not affected in Foxm1 −/− 9.5-dpc hearts (Fig. 3H). To determine whether Foxm1 deficiency would result in decreased entry of cardiomyocytes into mitosis, we performed immunofluorescent staining of heart sections with an antibody specific to the phosphorylated form of Histone 3 (PH3). Consistent with proliferation defects in Foxm1 −/− livers (Krupczak-Hollis et al., 2004), a 75% reduction in the number of cardiomyocytes undergoing mitosis was observed in Foxm1 −/− 15.5-dpc hearts (Fig. 3C,D,I). Interestingly, cell death did not contribute to reduced cardiomyocyte number in Foxm1 −/− 15.5-dpc hearts, as was determined by immunostaining with antibody against cleaved caspase 3 and TUNEL assay (Fig. 3E,F; and data not shown). These results demonstrated that Foxm1 deficiency reduces both DNA replication and mitosis in developing cardiomyocytes.

thumbnail image

Figure 3. Diminished DNA replication and mitosis in embryonic Foxm1 −/− cardiomyocytes. A,B:Foxm1 −/− hearts exhibited diminished number of BrdU-positive cardiomyocytes. Paraffin sections from BrdU-labeled WT (A) and Foxm1 −/− (B) 15.5-dpc embryos were stained with BrdU antibody followed by anti-mouse antibody conjugated with alkaline phosphatase and BCIP/NBT substrate. C,D:Foxm1 −/− 15.5-dpc hearts exhibited a decreased number of PH3-positive cardiomyocytes. E,F:Foxm1 deficiency did not cause an increase in apoptosis as demonstrated by immunostaining with antibody against cleaved caspase 3. G: Total number of cardiomyocytes is decreased in Foxm1 −/− hearts. Numbers of DAPI-stained cardiomyocyte nuclei were counted in ten random 400× microscope fields from Foxm1 −/− and WT 15.5-dpc hearts. H,I:Foxm1 −/− 15.5-dpc cardiomyocytes display diminished DNA replication and mitosis, whereas no proliferation defects were observed in Foxm1 −/− 9.5-dpc hearts. Numbers of BrdU (H) or PH3-positive cardiomyocytes (I) were counted in ten random 400× microscope fields. Mean ± S.D. was determined using three different WT or Foxm1 −/− 9.5-dpc and 15.5-dpc embryos. *P < 0.05. Magnification: ×200 (A–F).

Download figure to PowerPoint

Gene Expression Profile in Foxm1 −/− Heart

In order to determine potential target genes regulated by Foxm1 in the developing heart, we performed RT-PCR analysis of total heart RNA prepared from WT or Foxm1 −/− 14.5-dpc embryos. Although we confirmed undetectable levels of Foxm1 mRNA in Foxm1 −/− hearts (Fig. 4), Foxm1 −/− embryos displayed normal cardiac levels of Polo-like kinase 1, Aurora B kinase, Cyclin B1, Skp2, Cks1, and Lama4 (Fig. 4), all of which are known transcriptional targets for Foxm1 protein in hepatocytes and fibroblasts (Krupczak-Hollis et al., 2004; Wang et al., 2005). These results suggest that the Foxm1 deficiency does not influence expression of these genes in the developing heart. However, consistent with decreased entry into mitosis in Foxm1 −/− cardiomyocytes (Fig. 3I), we observed an 80% reduction in the expression of Cdc25B phosphatase (Fig. 4), a known transcriptional target for Foxm1 protein (Wang et al., 2001). Because Cdc25B is essential for activation of Cdk1/ cyclin B complex during M-phase progression (Borgne and Meijer, 1996), diminished Cdc25B levels may cause delayed cardiomyocyte entry into mitosis in Foxm1 −/− hearts, causing polyploid phenotype. Furthermore, numbers of cells expressing Cdk inhibitor p21Cip1 (p21) protein were increased in Foxm1 −/− hearts (Fig. 5A–E), a finding consistent with elevated p21 protein levels in Foxm1 −/− MEFs (Wang et al., 2005). Because p21 reduces Cdk2 activity, causing diminished DNA replication (Sherr and Roberts, 1999), induced cardiac p21 levels may contribute to the reduction of cell cycle progression in Foxm1 −/− hearts.

thumbnail image

Figure 4. Foxm1 −/− hearts displayed reduced expression of the nuclear factor of activated T-cell c3 (NFATc3), and the Cdk1-activator Cdc25B phosphatase. WT and Foxm1 −/− 14.5-dpc hearts were dissected and then used for preparation of total RNA. Semi-quantitative RT-PCR analysis was performed with primers specific to Foxm1, NFATc3, NFATc4, H11 kinase, 5-HT2B receptor, Calmodulin, FoxO3a, Pecam-1, Cdc25B, Aurora B, Plk-1, Cyclin B1, Skp2, Cks1, Laminin α4, Laminin α2, Integrin β1, and cyclophilin. Each individual sample was normalized to its corresponding cyclophilin level. Values are means ± S.D. *P < 0.05 is significant.

Download figure to PowerPoint

thumbnail image

Figure 5. Increased nuclear accumulation of p21 protein in embryonic Foxm1 −/− cardiomyocytes. A–D:Foxm1 −/− cardiomyocytes exhibit an increase in nuclear protein levels of Cdk inhibitor p21cip1. Heart paraffin sections from either Foxm1 −/− or WT 15.5-dpc embryos were stained with p21cip1 antibody followed by the secondary antibody conjugated with biotin, avidin-HRP, and DAB substrate (brown). Slides were counterstained with hematoxylin (blue). E:Foxm1 −/− hearts display increased numbers of p21cip1-positive cardiomyocytes (arrowheads in C,D). Numbers of p21cip1-positive cells in WT and Foxm1 −/− 15.5-dpc hearts were counted in ten random 400× microscope fields. Mean ± S.D. was determined using three different WT or Foxm1 −/− embryos. *P < 0.05 is significant. Magnifications: ×100 (A,B), ×400 (C,D).

Download figure to PowerPoint

Foxm1 −/− hearts displayed a 70% reduction in the expression levels of the nuclear factor of activated T-cell c3 (NFATc3), whereas cardiac expression of related NFATc4 protein was not changed in the absence of Foxm1 (Fig. 4). Since NFATc3 protein regulates genes essential for cardiac hypertrophy and cardiomyocyte proliferation (Wilkins et al., 2002; Sanna et al., 2005), diminished NFATc3 levels may contribute to the proliferation defects seen in Foxm1 −/− cardiomyocytes. Interestingly, we observed normal cardiac expression of calmodulin, H11 kinase, serotonin 5-HT2B receptor, and FoxO3a transcription factor (Fig. 4), all of which are essential for the development of hypertrophy in cardiomyocytes (Nebigil et al., 2000; Depre et al., 2002; Colomer et al., 2004; Skurk et al., 2005).

Endothelial-Specific Deletion of Foxm1 fl/fl Allele Does Not Alter Proliferation Rates in the Heart

Previous reports demonstrated that Foxm1 −/− embryos displayed diminished number of blood vessels in the liver and lung (Krupczak-Hollis et al., 2004; Kim et al., 2005a), suggesting vascular defects in these organs. Furthermore, endothelial-specific disruption of Foxm1 diminished proliferation of endothelial cells following LPC-induced vascular injury (Zhao et al., 2006). Since Foxm1 is expressed in both cardiomyocytes and cardiac endothelial cells in the developing heart (Fig. 2A–D), we sought to determine whether Foxm1 −/− embryos exhibit defects in development of cardiac endothelial cells. Immunohistochemical staining and RT-PCR analysis of 14.5 dpc Foxm1 −/− hearts displayed normal cardiac expression of the Platelet-endothelial cell adhesion molecule-1 (Pecam-1), suggesting no gross defects in vascular patterning (Fig. 4 and data not shown).

To determine whether the Foxm1 −/− cardiomyocyte defects were indirect due to defects in development of Foxm1-deficient endothelial cells, we used a Tie2-Cre recombinase transgene to mediate endothelial-specific deletion of the Foxm1 fl/fl allele in mouse embryos. Although the majority of Tie2-Cre Foxm1 fl/fl mice survived at birth, approximately 20% of Tie2-Cre Foxm1 fl/fl embryos died in late gestation-postnatally. These Tie2-Cre Foxm1 fl/fl embryos with the most severe phenotype exhibited vascular abnormalities in the lung as demonstrated by diminished Pecam-1 staining (Fig. 6I–J), a result consistent with reduced Pecam-1 levels in Foxm1 −/− lungs (Kim et al., 2005a). These same Tie2-Cre Foxm1 fl/fl embryos were then examined for heart defects. The Tie2-Cre Foxm1 fl/fl 17.5-dpc hearts displayed undetectable Foxm1 staining in cardiac endothelial cells (Fig. 6G,H), indicating efficient endothelial-specific deletion of Foxm1 fl/fl allele in the developing heart. However, no morphological defects were observed in the structure of coronary vessels and cardiac microcirculation in Tie2-Cre Foxm1 fl/fl heart (Fig. 6A,B and data not shown). Furthermore, the Tie-2 Foxm1 fl/fl hearts displayed normal structure of atria-ventricular valves and the ventricular outflow tract (Fig. 6C,D and data not shown), suggesting that Foxm1 is not essential for the formation of endocardial cushion during heart development.

thumbnail image

Figure 6. Endothelial Foxm1 deficiency causes normal proliferation rates in the heart. A–D: Tie2-Cre Foxm1 −/− embryos displayed normal structure of atrio-ventricular (A/V) valves (C,D) and myocardium (A,B). Foxm1 fl/fl and Tie2-Cre Foxm1 −/− 17.5-dpc hearts were fixed, paraffin-embedded, sectioned, and then H&E-stained. E–J: Heart (E–H) and lung paraffin sections (I,J) from Foxm1 fl/fl and Tie2-Cre Foxm1 −/− 17.5-dpc embryos were stained with antibodies specific to Foxm1 (G,H), PCNA (E,F), or Pecam-1 (I,J). Slides were counterstained with either nuclear fast red (red, E–H) or hematoxylin (blue, I,J). Tie2-Cre Foxm1 −/− hearts exhibited undetectable Foxm1 staining in endothelial cells (arrows in H; arrowheads show Foxm1 staining in proliferating cardiomyocytes) and normal PCNA staining (E,F). Tie2-Cre Foxm1 −/− lungs displayed diminished Pecam-1 staining (I,J). K: Total numbers of cardiomyocytes are similar in H&E-stained Tie2-Cre Foxm1 −/− and Foxm1 fl/fl 17.5-dpc hearts. L: Tie2-Cre Foxm1 −/− cardiomyocytes display normal DNA replication. Numbers of PCNA-positive cells were counted in ten random 400× microscope fields. Mean ± S.D. was determined using three Tie2-Cre Foxm1 −/− and Foxm1 fl/fl embryos. V, blood vessel; Br, bronchus; A/V, atrio-ventricular valve. Magnifications: ×100 (A–D,I,J), ×200 (E,F), and ×400 (G,H).

Download figure to PowerPoint

The Tie2-Cre Foxm1 fl/fl embryos displayed similar numbers of ventricular cardiomyocytes compared to control Foxm1 fl/fl embryos (Fig. 6K). Furthermore, cardiomyocyte proliferation was similar in Tie2-Cre Foxm1 fl/fl and Foxm1 fl/fl hearts (Fig. 6E,F,L), as demonstrated by immunohistochemical staining for Proliferation Cell Nuclear Antigen (PCNA). These results suggest that the endothelial Foxm1 deficiency does not contribute to proliferation defects in embryonic Foxm1 −/− heart.

Foxm1 Depletion in HL-1 Cardiomyocytes Causes Diminished DNA Replication, Delayed Entry Into Mitosis, and Reduced Expression of NFATc3 and Cdc25B

To determine whether Foxm1 deficiency directly reduces the cell cycle progression in cardiomyocytes, HL-1 mouse cardiomyocytes were transfected with short interfering RNA (siRNA) duplex specific to the mouse Foxm1 cDNA (siFoxm1) or with mutant siFoxm1 duplex (mutFoxm1) containing two mutations in the siRNA structure (Fig. 7D). Seventy-two hours later, cells were labeled for 2 hr with BrdU and then used for immunostaining with BrdU antibody. Depletion of Foxm1 caused a 50% reduction in the number of HL-1 cells undergoing DNA replication compared to either untransfected HL-1 cells or cells transfected with mutFoxm1 duplex (Fig. 7A,B). Furthermore, we performed time-course experiments in serum-starved HL-1 cells to determine whether Foxm1-deficiency influences mitotic progression in cardiomyocytes. BrdU and DAPI double staining demonstrated that although mitotic figures in control HL-1 cultures were detected at 16 hr following BrdU labeling, no mitotic figures were observed in siFoxm1-transfected HL-1 cells until 24 hr after BrdU labeling (Fig. 8), suggesting that Foxm1 deficiency causes an 8-hr delay in the entry of HL-1 cells into mitosis. These results suggest that Foxm1 is critical for both DNA replication and mitotic progression in HL-1 cardiomyocytes.

thumbnail image

Figure 7. Foxm1 depletion by siRNA causes diminished DNA replication and NFATc3 expression in HL-1 cardiomyocytes. A,B: siFoxm1 transfection diminishes the total number of cells undergoing DNA synthesis. HL-1 cells transfected with either siFoxm1 or mutant siFoxm1 for 72 hr were pulse-labeled with BrdU for 2 hr and then immunostained with BrdU antibodies (A). Numbers of BrdU-positive cells are presented as means ± S.D. from three distinct transfections (B). *P < 0.05. C: Quantitative real-time RT-PCR demonstrated that siFoxm1 transfection inhibits expression of Foxm1, Cdc25B and NFATc3 in HL-1 cardiomyocytes. D: Depletion of Foxm1 caused increased nuclear levels of p21 protein. Western blot analysis was performed using nuclear extracts from siRNA-transfected HL-1 cells. Cdk2 protein levels were used as loading controls. E: Schematic drawing of −10-Kb region of the mouse NFATc3 promoter with four potential Foxm1-binding sites. F: Chromatin Immunoprecipitation (ChIP) assay demonstrated that Foxm1 protein binds to the −9,259/−9,288 NFATc3 promoter region. Cross-linked chromatin from WT 17.5-dpc hearts (left panel) or CMV-Foxm1 transfected HL-1 cells (right panel) was used for immunoprecipitation (IP) with antibodies specific to Foxm1 or IgG control. The IP genomic DNA was analyzed for the amount of mouse NFATc3 promoter DNA using PCR analysis.

Download figure to PowerPoint

thumbnail image

Figure 8. Foxm1 depletion by siRNA causes delayed entry of HL-1 cells into mitosis. HL-1 cells transfected with either siFoxm1 or mutant siFoxm1 (control) for 24 hr were serum-starved for an additional 48 hr and then re-stimulated and pulse-labeled with BrdU for 2 hr. Immunofluorescent staining with BrdU antibodies (red) shows the presence of cells undergoing metaphase (white arrows) and telophase (yellow arrows). Cells were counterstained with DAPI (blue). Magnification is ×400.

Download figure to PowerPoint

Since Foxm1 −/− hearts displayed reduced expression of NFATc3 and Cdc25B phosphatase (Fig. 4), and increased nuclear levels of the Cdk-inhibitor p21cip1 protein (Fig. 5C,D), we examined the expression levels of these genes in siFoxm1-transfected HL-1 cardiomyocytes by Western blot and quantitative real-time RT-PCR analysis. These transfection studies revealed that siFoxm1 efficiently reduces the expression of mouse Foxm1 in HL-1 cells, whereas transfection of mutant siFoxm1 duplex has a minimal effect on Foxm1 levels (Fig. 7C). In comparison with either untransfected cells or cells transfected with mutant siFoxm1, depletion of Foxm1 levels by siFoxm1 caused a significant decrease in the expression of M-phase promoting Cdc25B phosphatase and NFATc3 transcription factor (Fig. 7C). Furthermore, Western blot demonstrated that transfection of siFoxm1 caused increased nuclear levels of Cdk inhibitor p21cip1 protein (Fig. 7D), a result consistent with increased cardiac staining for p21cip1 protein in Foxm1 −/− hearts (Fig. 5C,D). Interestingly, expression levels of Cyclin B1 and Cdk-inhibitor p27Kip1 were not affected by siFoxm1 transfection (Fig. 7D), suggesting that Foxm1-deficiency does not influence the expression of these genes in cultured HL-1 cells.

Foxm1 Binds to the −9,259/−9,288-bp Region of Endogenous Mouse NFATc3 Promoter

Previous studies demonstrated that Foxm1 directly regulates Cdc25B and p21Cip1 genes in hepatocytes and mouse embryonic fibroblasts (Wang et al., 2001; Krupczak-Hollis et al., 2004; Wang et al., 2005). However, little is known about cardiomyocyte-specific Foxm1 targets. Since NFATc3 mRNA levels were reduced in both Foxm1 −/− hearts and Foxm1-depleted HL-1 cardiomyocytes, we used Chromatin Immunoprecipitation (ChIP) assays to determine whether Foxm1 protein directly binds to the NFATc3 promoter in context of embryonic mouse DNA and to identify specific Foxm1-binding regions. Four potential Foxm1 DNA binding sites were found in the −10-Kb promoter region of the mouse NFATc3 gene: −4,263/−4,277, −4,989/−5,000, −5,533/−5,547, and −9,259/−9,288 bp, the latter of which contains eight overlapping Foxm1-binding motifs (Fig. 7E). The cross-linked and sonicated chromatin from mouse WT 17.5-dpc embryonic hearts was immunoprecipitated (IP) with antibodies specific to either Foxm1 or IgG control antibody. Binding of NFATc3 promoter DNA associated with the IP chromatin was determined by PCR with primers specific to potential Foxm1-binding sites in mouse NFATc3 promoter. Although Foxm1 did not recognize the first three NFATc3 promoter regions (data not shown), Foxm1 specifically binds to the −9,259/−9,288-bp region of endogenous mouse NFATc3 promoter (Fig. 7F, left panel). Furthermore, mouse HL-1 cardiomyocytes transfected with CMV-Foxm1 expression vector displayed efficient binding of the −9,259/−9,288 bp NFATc3 promoter region and Foxm1 protein (Fig. 7F, right panel). These results suggest that the mouse NFATc3 gene is a direct transcriptional target of Foxm1 in cardiomyocytes, a result consistent with reduced NFATc3 levels in Foxm1 −/− heart (Fig. 4) and Foxm1-depleted HL-1 cardiomyocytes (Fig. 7C).

DISCUSSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES

The requirement for Foxm1 activity in cardiac development was examined in Foxm1 −/− mouse embryos that contain a targeted deletion of exons 4 through 7, encoding the Foxm1 DNA binding domain and C-terminal transcriptional activation domains, both of which are essential for Foxm1 transcriptional activity (Krupczak-Hollis et al., 2004). Most of the Foxm1 −/− embryos exhibited an embryonic lethal phenotype between 13.5 and 16.5 dpc because of severe abnormalities in liver morphogenesis, hypertrophy of blood vessels, and inability of lung mesenchyme to form peripheral pulmonary capillaries (Krupczak-Hollis et al., 2004; Kim et al., 2005a). Overall developmental defects in our Foxm1 −/− embryos were more severe compared to the previously reported Foxm1−/−neo mouse line, which contains a targeted insertion of PGK-Neo into Foxm1 gene locus without deleting any of the coding sequences (Korver et al., 1998). This includes a complete embryonic lethality, earlier thinning of myocardium, and dilation of the heart chambers (Krupczak-Hollis et al., 2004; this report). One explanation for these differences between Foxm1 −/− and Foxm1−/−neo mouse lines is that the Foxm1−/−neo mice may contain hypomorphic levels of Foxm1 protein, providing residual Foxm1 activity during embryonic development. This conclusion is supported by the fact that Foxm1 −/− mouse embryonic fibroblasts (MEF) were unable to grow in culture due to severe mitotic defects and premature senescence (Wang et al., 2005), whereas Foxm1−/−neo MEFs grew at passage 4 in culture, displaying reduced G2/M progression and abnormal mitosis due to defective chromosomal segregation (Laoukili et al., 2005).

Proliferation defects in Foxm1 −/− cardiomyocytes were also associated with increased nuclear expression of p21Cip1 protein (p21), which inactivates the Cdk2 protein required for DNA replication (Sherr and Roberts, 1999). Cdk2 complexes with either Cyclin E or Cyclin A cooperate with Cyclin D-Cdk4/6 to phosphorylate the Retinoblastoma (RB) protein, which releases bound E2F transcription factor and allows it to stimulate expression of genes required for DNA replication (Harbour and Dean, 2000; Ishida et al., 2001). Therefore, defects in DNA replication in Foxm1 −/− hearts may be a direct consequence of increased cardiac levels of p21 protein. This result is consistent with previously published data demonstrating that transgenic over-expression of Foxm1 in Rosa26-Foxm1 transgenic mice is associated with decreased p21 protein levels and premature proliferation of different lung cell types following Butylated Hydroxytoluene (BHT) lung injury (Kalinichenko et al., 2003). We previously demonstrated that Foxm1 regulates transcription of Skp2 and Cks1 proteins, which are specificity subunits of the Skp1-Cullin-F-Box (SCF) ubiquitin ligase complex that targets both p21 and p27 proteins for degradation during the G1/S transition (Wang et al., 2005). Unchanged expression levels of Skp2 and Cks1 in Foxm1 −/− hearts suggest that degradation of p21 protein by SCF ubiquitin ligase is not involved in abnormal nuclear accumulation of p21 protein in the heart. Interestingly, accumulation of polyploid cells was found in livers and lungs of Skp2 −/− mice, but not in Skp2 −/− hearts (Kossatz et al., 2004), suggesting that the Skp2 protein is not essential for cardiomyocyte proliferation. Although the mechanism of nuclear p21 accumulation in Foxm1 −/− cardiomyocytes remains unknown, cardiac p21 mRNA levels remained unchanged in Foxm1 −/− hearts (data not shown), excluding a transcriptional regulation of p21 gene by the Foxm1.

Proliferation defects in Foxm1 −/− hepatoblasts and vascular smooth muscle cells were associated with reduced expression of Polo-like kinase 1 (PLK1) and Aurora B kinase (Krupczak-Hollis et al., 2004; Kim et al., 2005a), which contribute to a failure to complete mitosis and cytokinesis. (Glover et al., 1998; Adams et al., 2001). Furthermore, a recent study demonstrated that these genes are direct transcriptional targets for the Foxm1 transcription factor in MEFs and U2OS human osteosarcoma cell line (Wang et al., 2005). Interestingly, even though cardiac expression of PLK1 and Aurora B kinase was unchanged in embryonic Foxm1 −/− hearts, Foxm1 −/− cardiomyocytes still exhibited reduced mitosis causing the accumulation of polyploid cells. These results suggest that Foxm1 may regulate other genes essential for cardiomyocyte entry into mitosis during heart morphogenesis. In this study, we found that Foxm1 −/− hearts display reduced expression of Cdc25B phosphatase, a result consistent with diminished levels of Cdc25B in proliferating Foxm1 −/− hepatocytes following a partial hepatoctomy (Wang et al., 2001). In addition, increased nuclear levels of Cdk inhibitor p21 may not only prevent activation of Cdk2, but p21 can also complex with and inhibit activity of M-phase promoting Cdk1 protein (Op De Beeck et al., 2001; Patel et al., 2002; Dash and El-Deiry, 2005). Furthermore, reduced Cdc25B expression and increased nuclear p21 protein levels were found in cultured HL-1 cardiomyocytes after siRNA-mediated Foxm1 depletion. Progression into mitosis requires the activation of Cdk1 through assembly with cyclin B1 regulatory subunit and the removal of Cdk1 inhibitory phosphates at Thr 14 and Tyr 15 by the Cdc25B and Cdc25C phosphatases (Borgne and Meijer, 1996; Wells et al., 1999; Nilsson and Hoffmann, 2000). Therefore, diminished cardiac levels of Cdc25B and increased nuclear levels of p21 protein can decrease Cdk1 activation and M-phase progression causing delayed entry into mitosis in Foxm1 −/− cardiomyocytes.

The nuclear factor of the activated T-cell (NFAT) family of transcription factors regulates genes essential for morphogenesis of the heart, skeletal muscle, cartilage, bone, skin, neuronal, and immune systems (reviewed in Graef et al., 2001). NFATc3 activity is critical for the generation of primary fibers in skeletal muscle (Crabtree and Olson, 2002), whereas over-expression of NFATc4 induces cardiac hypertrophy in adult mice (Molkentin, 2000). In this study, we demonstrated that Foxm1 −/− embryos display a significant reduction in the expression of NFATc3 in the heart, whereas NFATc4 levels remained unchanged. Since a combinatorial disruption of NFATc3 and NFATc4 genes caused embryonic lethality and diminished cardiomyocyte proliferation (Bushdid et al., 2003), it is unlikely that NFATc3 deficiency is solely responsible for the proliferation defects in Foxm1 −/− cardiomyocytes. However, in combination with either reduced Cdc25B or elevated p21 levels, diminished NFATc3 expression may contribute to the cell cycle defects in developing Foxm1 −/− heart. Interestingly, quantitative real-time RT-PCR demonstrated that siRNA-mediated Foxm1 depletion in cultured HL-1 cardiomyocytes causes a significant decrease in the expression of NFATc3, suggesting that the mouse NFATc3 gene is a direct transcriptional target of Foxm1 in cardiomyocytes. Furthermore, we used Chromatin Immunoprecipitation (ChIP) assay to demonstrate that Foxm1 protein directly binds to the −9,259/−9,288-bp region of endogenous mouse NFATc3 promoter in the developing heart and in cultured HL-1 cells, providing a further support for this concept.

Previous studies have demonstrated that endothelial-specific disruption of either neurofibromin (NF1) or platelet-derived growth factor-B (PDGFB) caused abnormal development of myocardium (Gitler et al., 2003; Bjarnegard et al., 2004). These studies suggest that signaling from cardiac endothelial cells is required for heart development. Since Foxm1 is expressed in both cardiomyocytes and cardiac endothelial cells (Ye et al., 1997; and this report), we used Tie2-Cre Foxm1 fl/fl mice to investigate the consequences of endothelial Foxm1 deficiency for heart development. Surprisingly, the Tie2-Cre Foxm1 fl/fl embryos displayed normal heart morphology and similar cardiomyocyte proliferation compared to control Foxm1 fl/fl embryos. These results suggest that the endothelial Foxm1 deficiency does not contribute to proliferation defects in embryonic Foxm1 −/− heart. Although the Tie2-Cre Foxm1 fl/fl embryos exhibited defects in development of pulmonary vasculature, no gross morphological defects in vascular patterning were observed in Tie2-Cre Foxm1 −/− hearts. This indicates that Foxm1 promotes the lung vascular development, but is dispensable for the development of cardiac endothelium. This is consistent with previous reports demonstrating that Foxm1 −/− embryos displayed reduced numbers of endothelial cells in developing lung (Krupczak-Hollis et al., 2004; Kim et al., 2005a), suggesting different requirements for Foxm1 activity in distinct populations of endothelial cells.

In summary, Foxm1 −/− embryos displayed ventricular hypoplasia and severe reduction in cardiomyocyte DNA replication and mitosis. Proliferation defects in Foxm1 −/− cardiomyocytes were associated with reduced expression of NFATc3, Cdc25B phosphatase, and abnormal nuclear accumulation of the Cdk-inhibitor p21cip1 protein. Depletion of Foxm1 expression by short interfering RNA caused altered expression of these genes in cultured HL-1 cardiomyocytes. Endothelial-specific deletion of the Foxm1 fl/fl allele in mouse embryos did not affect heart development and cardiomyocyte proliferation. Foxm1 binds to the −9,259/−9,288-bp region of endogenous mouse NFATc3 promoter in the developing heart, suggesting that the mouse NFATc3 gene is a direct transcriptional target of Foxm1 during heart development. The identification of critical regulators of cardiomyocyte proliferation, such as Foxm1, may provide novel strategies for diagnosis and treatment of congenital heart defects and dilated cardiomyopathy.

EXPERIMENTAL PROCEDURES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES

Foxm1 −/− Mice

We previously described the generation of Foxm1 +/− mice, in which the targeted allele lacked the DNA binding and transcriptional activation domains (Krupczak-Hollis et al., 2004). These Foxm1 +/− mice were bred for eight generations into the C57BL/6 mouse genetic background. Pregnant females were killed by carbon dioxide asphyxiation to dissect Foxm1 −/− embryos at various days of gestation. Internal cartilage tissue samples were used for genotyping the Foxm1 null allele by PCR analysis as described previously (Krupczak-Hollis et al., 2004). Two hours before sacrifice, an i.p. injection of PBS containing 10 mg/ml 5-bromo-2′-deoxyuridine (BrdU, Sigma; 50 μg/g body weight) was administered to pregnant females.

Conditional knockout mice containing an endothelial-specific deletion of the Foxm1 LoxP allele were generated by breeding Foxm1 LoxP/LoxP (fl/fl) mice with transgenic Tie2-Cre C57BL/6 mice (Zhao et al., 2006). Although 20% of Tie2-Cre Foxm1 fl/fl embryos died in late gestation/postnatally, the majority of Tie2-Cre Foxm1 fl/fl mice survived through birth and were capable of mating (Zhao et al., 2006).

Immunohistochemical Staining

Wild type (WT), Foxm1 −/−, Tie2-Cre Foxm1 fl/fl, and Foxm1 fl/fl embryos were harvested, fixed overnight with 10% buffered formalin, and then embedded into paraffin blocks. Paraffin in 5-μm sections was stained with hematoxylin and eosin (H&E) for morphological examination and morphometric analysis. The thickness of myocardial walls and heart trabeculae, and the length of trabeculae were measured on photomicrographs using Axiovision 3.0 software. Means ± SD were calculated from the average measurements from three distinct Foxm1 −/− and WT embryos, in which 10 randomized sections of left ventricle were analyzed for each genotype. The heart size was determined as the largest size between apex and the tip of right atrium using 10 heart sections for each embryo (five embryos per group). The thickness of atrioventricular (A/V) valve leaflets was determined from an average of measurements taken at the base, middle, and tip of the tricuspid and mitral valve leaflets. An overall average was calculated from 7 random heart sections from three embryos for each genotype. Statistical significance was determined using a two-tailed Student's t-test of values for either Foxm1 −/− vs. WT hearts or Tie2-Cre Foxm1 fl/fl vs. Foxm1 fl/fl hearts (P < 0.05).

Paraffin sections were used for immunostaining with mouse monoclonal antibodies against desmin (1:1,000; A8; Sigma, St Louis, MO), BrdU (1:100; Bu 20A; Dako), and PCNA (1:1,000; PC-10; Roche), rabbit polyclonal antibodies against Foxm1 (1:100; (Kalinichenko et al., 2004)), and cleaved caspase 3 (1:200; 5A1; Cell Signaling) or goat polyclonal antibodies against laminin α4 (1:100; V-20; Santa Cruz). Antibody-antigen complexes were detected using either secondary antibody directly conjugated with alkaline phosphatase (AP) or biotinylated secondary antibody followed by avidin-AP complex (Vector Labs, Burlingame, CA). BCIP/NBT was used as AP-substrate as described (Kalinichenko et al., 2001b). Sections were counterstained with nuclear fast red (Vector Labs). We also used mouse monoclonal antibodies against p21Cip1 (1:200; SXM30; BD Biosciences), goat polyclonal antibodies against GATA-4 (1:100; C-20; Santa Cruz), and rat monoclonal antibodies against Pecam-1 (1:50; MEC 13.3, Pharmingen). Antibody-antigen complexes were detected using biotinylated secondary antibody, avidin-horseradish peroxidase (HRP) complex, and DAB substrate (all from Vector Labs) as described (Kalinichenko et al., 2002; Kim et al., 2005b). For immunofluorescent detection of phosphohistone H3 (PH3), we used rabbit polyclonal PH3 antibody (1:200; MC463; Upstate) followed by anti-rabbit antibody conjugated with FITC (Vector Labs) as described previously (Kalinichenko et al., 2003).

Semi-Quantitative RT-PCR and Quantitative Real-Time RT-PCR

Total mouse heart RNA was prepared from 14.5-dpc hearts of WT or Foxm1 −/− embryos using RNA-STAT-60 (Tel-Test “B” Inc. Friendswood, TX) and then used for reverse transcriptase (RT)-PCR analysis as described (Kalinichenko et al., 2002). The following sense and antisense primers were used for amplification: laminin α4 (Lama4), 5′- ggatcccgactggtcattgatggtctacgag and 5′-gtcgaccgccttctgtggaaaaataagttc; Foxm1, 5′-ggatcctgccaccccagaccttgttc and 5′- gtcgactccct- gatgcttttcgctgtc; Cdc25B, 5′-attccag- ctctgcccaagctttggc and 5′-tccacaaatccgtcatcttctca; Aurora B, 5′-ttgacaactttgagattggg and 5′-gctggtcgtagaagtagttgt; Plk-1, 5′- ctcctggagctgcacaaga- ggaggaa and 5′- tctgtctgaagcatcttctggatgag; Cyclin B1, 5′-atcggggaacctctgatttt and 5′-tcacacacaggcaccttctc; Skp2, 5′- gtatgttagggaaccatttgcgag and 5′-ttagaagggcacttggaagagtt; Cks1, 5′-gacctcaaagccctcgtgt and 5′- tgaaacataaatccataagtcatca; laminin α2 (Lama2), 5′-tgtcgtggggattctgtatgtc and 5′-caagaaggtccaatccaacttt; integrin β1, 5′-attggctttggctcatttgtgg and 5′-ccagcagtcgtgttacattcc; NFATc3, 5′-tggatctcagtatc- ctttaa and 5′-cacacgaaatacaagtcgga; NFATc4, 5′-cattggcactgcagatgag and 5′- cgtagctcaatgtctgaat; H11 Kinase, 5′-cttgcttggttgcgttaggt and 5′-tggtggtgatggtttgagag; 5-HT2B receptor, 5′-ctcttttcaactgcctccatc and 5′-ccagcattgcc- accttttc; calmodulin, 5′-tccgttcttccttcttcgctcgcaccatggc and 5′-gtgtgtggacagaggggcttctgacatcag; FoxO3a, 5′-gtcacactacggcaaccaga and 5′-cctgagagaga- gtccgagagg; cyclophilin, 5′-agctctgagcactggagagaaa and 5′-tcctgagctacagaaggaatgg. Two different cDNA concentrations were used for RT-PCR reactions to ensure that RT-PCR conditions were in the linear range. Quantitation of expression levels was determined with Tiff files of ethidium bromide-stained gels by using the BioMax 1D program (Kodak) as described (Kalinichenko et al., 2002).

For the real-time RT-PCR, the BioRAD cDNA Synthesis Kit containing both oligo-dT and random hexamer primers was used to synthesize cDNA from 10 μg of total RNA prepared from mouse HL-1 cells as described (Wang et al., 2005; Kim et al., 2006). The following sense and anti-sense primers and annealing temperature (Ta) were used to amplify and measure the amount of mRNA by Real-Time RT-PCR: Foxm1, 5′-cacttggattgaggaccactt and 5′-gtcgtttctgctgtgattcc (Ta: 57.5°C); Cdc25B, 5′-ccctt- ccctgttttcctttc and 5′-acacacactcctgccatagg (Ta: 61.7°C), NFATc3, 5′-actgcctcatcaccatctcc and 5′-tcccaataatctcgttcacatc (Ta: 58.0°C), and cyclophilin, 5′-ggcaaatgctggaccaaacac and 5′- ttcctggacccaaaacgctc (Ta: 57.5°C). Reactions were amplified and analyzed in triplicate using a MyiQ Single Color Real-Time PCR Detection System (Bio-Rad, CA).

siRNA Transfection

Mouse cardiomyocyte HL-1 cells were obtained from Dr. Claycomb and were cultured in Claycomb medium (JRH Bioscience). In order to inhibit Foxm1 expression in HL-1 cells, we used a 19-nucleotide short interfering RNA (siRNA) duplex specific to mouse Foxm1 cDNA (siFoxm1, 5′-gga cca cuu ccc uua cuu u-3′). Foxm1 siRNA containing symmetric 2-Uracil (U) 3′ overhangs was designed and synthesized using Dharmacon Research algorithm. We transfected 100 nM of either siFoxm1 or mutant siFoxm1 (mutFoxm1; mutated nucleotides are indicated by capital letters, 5′-gga cca cuu Ucc Cua cuu u-3′) duplexes into mouse HL-1 cells using Lipofectamine™ 2000 reagent (Invitrogen) in serum free tissue culture media as described previously (Wang et al., 2005; Kim et al., 2006). HL-1 cells were harvested at 72 hr after transfection for total RNA preparation or immunostaining. HL-1 cells were labeled with BrdU for 2 hr, and then fixed with ethanol and immunostained for BrdU incorporation using the BrdU-Labeling and Detection kit II (Roche Diagnostics) according to manufacturer's recommendations. For time-course experiments, HL-1 cells were transfected with siRNA. Twenty-four hours after transfection, the cells were serum starved for 48 hr and then re-stimulated and labeled with BrdU for 2 hr as described (Major et al., 2004).

Western Blot Analysis

Nuclear protein extracts were prepared from HL-1 cells, and then subjected to Western Blot analysis (Kalinichenko et al., 2001a) using mouse monoclonal antibody against cyclin B1 (1:500; GNS-11; BD Biosciences), p27Kip1 (1:3,000; G173-524; BD Biosciences), or p21Cip1 (1:200; SXM30; BD Biosciences). We also used rabbit polyclonal antibody against Cdk2 (1:1,500; M2; Santa Cruz) for loading controls. Detection of the immune complex was accomplished by using secondary antibodies directly conjugated with HRP followed by chemi-illuminescence (Supersignal, Pierce, Rockford, IL).

Chromatin Immunoprecipitation (ChIP) Assays

Embryonic heart tissue from WT 17.5-dpc mouse embryos was cross-linked by addition of formaldehyde and used to prepare protein extract as described (Wells and Farnham, 2002). Cross-linked extracts were also prepared from HL-1 cells that were transiently transfected with CMV-Foxm1 expression vector for 24 hr. Extracts were subsequently sonicated and used for the immunoprecipitation (IP) with Foxm1 rabbit antiserum as described previously (Rausa et al., 2003). IP with P-selectin rabbit antiserum (BD Biosciences, Palo Alto, CA) was used as a control. Crosslinks were reversed on all samples by addition of TE buffer containing 10 μg of RNase A and then incubated for 15 min at 25°C. Proteinase K (10 μg) was then added and samples were digested for 16 hr at 65°C. DNA was extracted from the digested samples using PCR purification columns according to the manufacturer's instructions (Qiagen, Maryland). We then used these ChIP DNA samples for PCR reaction using primers specific to the −9,259/−9,288-bp region of endogenous mouse NFATc3 promoter: sense 5′-ttgatgacctgaactccttgcc (−9,331/−9,310) and antisense 5′-tcttagaaaagtatgcccccacc (−9,230/−9,252).

Statistical Analysis

The Student's t-test was used to determine statistical significance. P values ≤ 0.05 were considered significant. Values for all measurements were expressed as the mean ± standard deviation (SD).

Acknowledgements

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES

We thank H. M. Yoder and A.V. Kalin for excellent technical assistance, and Eric Svensson for critically reviewing the manuscript. This work was supported by grants from National Institute of Health (HL 84151-01 to V.V.K.), American Heart Association ((0335036N to V.V.K.), and March of Dimes Birth Defects Foundation (6-FY2005-325 to V.V.K.).

REFERENCES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES
  • Adams RR, Carmena M, Earnshaw WC. 2001. Chromosomal passengers and the (aurora) ABCs of mitosis. Trends Cell Biol 11: 4954.
  • Bjarnegard M, Enge M, Norlin J, Gustafsdottir S, Fredriksson S, Abramsson A, Takemoto M, Gustafsson E, Fassler R, Betsholtz C. 2004. Endothelium-specific ablation of PDGFB leads to pericyte loss and glomerular, cardiac and placental abnormalities. Development. 131: 18471857.
  • Borgne A, Meijer L. 1996. Sequential dephosphorylation of p34(cdc2) on Thr-14 and Tyr-15 at the prophase/metaphase transition. J Biol Chem 271: 2784727854.
  • Bushdid PB, Osinska H, Waclaw RR, Molkentin JD, Yutzey KE. 2003. NFATc3 and NFATc4 are required for cardiac development and mitochondrial function. Circ Res 92: 13051313.
  • Clark KL, Halay ED, Lai E, Burley SK. 1993. Co-crystal structure of the HNF-3/fork head DNA-recognition motif resembles histone H5. Nature 364: 412420.
  • Clark KL, Yutzey KE, Benson DW. 2006. Transcription factors and congenital heart defects. Annu Rev Physiol 68: 97121.
  • Clevidence DE, Overdier DG, Tao W, Qian X, Pani L, Lai E, Costa RH. 1993. Identification of nine tissue-specific transcription factors of the hepatocyte nuclear factor 3/forkhead DNA-binding-domain family. Proc Natl Acad Sci USA 90: 39483952.
  • Colomer JM, Terasawa M, Means AR. 2004. Targeted expression of calmodulin increases ventricular cardiomyocyte proliferation and deoxyribonucleic acid synthesis during mouse development. Endocrinology 145: 13561366.
  • Crabtree GR, Olson EN. 2002. NFAT signaling: choreographing the social lives of cells. Cell 109: S67S79.
  • Cripps RM, Olson EN. 2002. Control of cardiac development by an evolutionarily conserved transcriptional network. Dev Biol 246: 1428.
  • Dash BC, El-Deiry WS. 2005. Phosphorylation of p21 in G2/M promotes cyclin B-Cdc2 kinase activity. Mol Cell Biol 25: 33643387.
  • Depre C, Hase M, Gaussin V, Zajac A, Wang L, Hittinger L, Ghaleh B, Yu X, Kudej RK, Wagner T, Sadoshima J, Vatner SF. 2002. H11 kinase is a novel mediator of myocardial hypertrophy in vivo. Circ Res 91: 10071014.
  • Eisenberg LM, Markwald RR. 2004. Cellular recruitment and the development of the myocardium. Dev Biol. 274: 225232.
  • Frey N, Olson EN. 2003. Cardiac hypertrophy: the good, the bad, and the ugly. Annu Rev Physiol 65: 4579.
  • Gitler AD, Zhu Y, Ismat FA, Lu MM, Yamauchi Y, Parada LF, Epstein JA. 2003. Nf1 has an essential role in endothelial cells. Nat Genet 33: 7579.
  • Glover DM, Hagan IM, Tavares AA. 1998. Polo-like kinases: a team that plays throughout mitosis. Genes Dev 12: 37773787.
  • Graef IA, Chen F, Crabtree GR. 2001. NFAT signaling in vertebrate development. Curr Opin Genet Dev 11: 505512.
  • Harbour JW, Dean DC. 2000. The Rb/E2F pathway: expanding roles and emerging paradigms. Genes Dev 14: 23932409.
  • Ishida S, Huang E, Zuzan H, Spang R, Leone G, West M, Nevins JR. 2001. Role for E2F in control of both DNA replication and mitotic functions as revealed from DNA microarray analysis. Mol Cell Biol 21: 46844699.
  • Kaestner KH, Lee KH, Schlondorff J, Hiemisch H, Monaghan AP, Schutz G. 1993. Six members of the mouse forkhead gene family are developmentally regulated. Proc Natl Acad Sci USA 90: 76287631.
  • Kalin TV, Wang IC, Ackerson TJ, Major ML, Detrisac CJ, Kalinichenko VV, Lyubimov A, Costa RH. 2006. Increased levels of the FoxM1 transcription factor accelerate development and progression of prostate carcinomas in both TRAMP and LADY transgenic mice. Cancer Res 66: 17121720.
  • Kalinichenko VV, Lim L, Beer-Stoltz D, Shin B, Rausa FM, Clark J, Whitsett JA, Watkins SC, Costa RH. 2001a. Defects in pulmonary vasculature and perinatal lung hemorrhage in mice heterozygous null for the Forkhead Box f1 transcription factor. Dev Biol 235: 489506.
  • Kalinichenko VV, Lim L, Shin B, Costa RH. 2001b. Differential expression of forkhead box transcription factors following butylated hydroxytoluene lung injury. Am J Physiol Lung Cell Mol Physiol 280: L695L704.
  • Kalinichenko VV, Zhou Y, Bhattacharyya D, Kim W, Shin B, Bambal K, Costa RH. 2002. Haploinsufficiency of the mouse Forkhead Box f1 gene causes defects in gall bladder development. J Biol Chem 277: 1236912374.
  • Kalinichenko VV, Gusarova GA, Tan Y, Wang IC, Major ML, Wang X, Yoder HM, Costa RH. 2003. Ubiquitous expression of the forkhead box M1B transgene accelerates proliferation of distinct pulmonary cell-types following lung injury. J Biol Chem 278: 3788837894.
  • Kalinichenko VV, Major M, Wang X, Petrovic V, Kuechle J, Yoder HM, Shin B, Datta A, Raychaudhuri P, Costa RH. 2004. Forkhead Box m1b transcription factor is essential for development of hepatocellular carcinomas and is negatively regulated by the p19ARF tumor suppressor. Genes Dev 18: 830850.
  • Kim IM, Ackerson T, Ramakrishna S, Tretiakova M, Wang IC, Kalin TV, Major ML, Gusarova GA, Yoder HM, Costa RH, Kalinichenko VV. 2006. The Forkhead Box m1 transcription factor stimulates the proliferation of tumor cells during development of lung cancer. Cancer Res 66: 21532161.
  • Kim IM, Ramakrishna S, Gusarova GA, Yoder HM, Costa RH, Kalinichenko VV. 2005a. The forkhead box M1 transcription factor is essential for embryonic development of pulmonary vasculature. J Biol Chem 280: 2227822286.
  • Kim IM, Zhou Y, Ramakrishna S, Hughes DE, Solway J, Costa RH, Kalinichenko VV. 2005b. Functional characterization of evolutionary conserved DNA regions in forkhead box f1 gene locus. J Biol Chem 280: 3790837916.
  • Korver W, Roose J, Clevers H. 1997. The winged-helix transcription factor Trident is expressed in cycling cells. Nucleic Acids Res 25: 17151719.
  • Korver W, Schilham MW, Moerer P, van den Hoff MJ, Dam K, Lamers WH, Medema RH, Clevers H. 1998. Uncoupling of S phase and mitosis in cardiomyocytes and hepatocytes lacking the winged-helix transcription factor trident. Curr Biol 8: 13271330.
  • Kossatz U, Dietrich N, Zender L, Buer J, Manns MP, Malek NP. 2004. Skp2-dependent degradation of p27kip1 is essential for cell cycle progression. Genes Dev 18: 26022607.
  • Krupczak-Hollis K, Wang X, Kalinichenko VV, Gusarova GA, Wang I-C, Dennewitz MB, Yoder HM, Kiyokawa H, Kaestner KH, Costa RH. 2004. The mouse Forkhead Box m1 transcription factor is essential for hepatoblast mitosis and development of intrahepatic bile ducts and vessels during liver morphogenesis. Dev Biol 276: 7488.
  • Laoukili J, Kooistra MR, Bras A, Kauw J, Kerkhoven RM, Morrison A, Clevers H, Medema RH. 2005. FoxM1 is required for execution of the mitotic programme and chromosome stability. Nat Cell Biol 7: 126136.
  • Li S, Zhou D, Lu MM, Morrisey EE. 2004. Advanced cardiac morphogenesis does not require heart tube fusion. Science 305: 16191622.
  • Lin AC, Roche AE, Wilk J, Svensson EC. 2004. The N termini of Friend of GATA (FOG) proteins define a novel transcriptional repression motif and a superfamily of transcriptional repressors. J Biol Chem 279: 5501755023.
  • Major ML, Lepe R, Costa RH. 2004. Forkhead Box M1B (FoxM1B) transcriptional activity requires binding of Cdk/Cyclin complexes for phosphorylation-dependent recruitment of p300/CBP co-activators. Mol Cell Biol 24: 26492661.
  • McFadden DG, Barbosa AC, Richardson JA, Schneider MD, Srivastava D, Olson EN. 2005. The Hand1 and Hand2 transcription factors regulate expansion of the embryonic cardiac ventricles in a gene dosage-dependent manner. Development 132: 189201.
  • Molkentin JD. 2000. Calcineurin and beyond: cardiac hypertrophic signaling. Circ Res 87: 731738.
  • Nebigil CG, Choi DS, Dierich A, Hickel P, Le Meur M, Messaddeq N, Launay JM, Maroteaux L. 2000. Serotonin 2B receptor is required for heart development. Proc Natl Acad Sci USA 97: 95089513.
  • Nilsson I, Hoffmann I. 2000. Cell cycle regulation by the Cdc25 phosphatase family. Prog Cell Cycle Res 4: 107114.
  • Olson EN. 2004. A decade of discoveries in cardiac biology. Nat Med 10: 467474.
  • Op De Beeck A, Sobczak-Thepot J, Sirma H, Bourgain F, Brechot C, Caillet-Fauquet P. 2001. NS1- and minute virus of mice-induced cell cycle arrest: involvement of p53 and p21(cip1). J Virol 75: 1107111078.
  • Patel V, Lahusen T, Sy T, Sausville EA, Gutkind JS, Senderowicz AM. 2002. Perifosine, a novel alkylphospholipid, induces p21(WAF1) expression in squamous carcinoma cells through a p53-independent pathway, leading to loss in cyclin-dependent kinase activity and cell cycle arrest. Cancer Res 62: 14011409.
  • Plageman TF Jr, Yutzey KE. 2005. T-box genes and heart development: putting the “T” in heart. Dev Dyn 232: 1120.
  • Rausa F, Tan Y, Costa RH. 2003. Association between HNF-6 and FoxA2 DNA binding domains stimulates FoxA2 transcriptional activity but inhibits HNF-6 DNA binding. Mol Cell Biol 23: 437449.
  • Sanna B, Bueno OF, Dai YS, Wilkins BJ, Molkentin JD. 2005. Direct and indirect interactions between calcineurin-NFAT and MEK1-extracellular signal-regulated kinase 1/2 signaling pathways regulate cardiac gene expression and cellular growth. Mol Cell Biol 25: 865878.
  • Sherr CJ, Roberts JM. 1999. CDK inhibitors: positive and negative regulators of G1-phase progression. Genes Dev 13: 15011512.
  • Skurk C, Izumiya Y, Maatz H, Razeghi P, Shiojima I, Sandri M, Sato K, Zeng L, Schiekofer S, Pimentel D, Lecker S, Taegtmeyer H, Goldberg AL, Walsh K. 2005. The FOXO3a transcription factor regulates cardiac myocyte size downstream of AKT signaling. J Biol Chem 280: 2081420823.
  • Svensson EC, Huggins GS, Lin H, Clendenin C, Jiang F, Tufts R, Dardik FB, Leiden JM. 2000. A syndrome of tricuspid atresia in mice with a targeted mutation of the gene encoding Fog-2. Nat Genet 25: 353356.
  • Wang IC, Chen YJ, Hughes D, Petrovic V, Major ML, Park HJ, Tan Y, Ackerson T, Costa RH. 2005. Forkhead box M1 regulates the transcriptional network of genes essential for mitotic progression and genes encoding the SCF (Skp2-Cks1) ubiquitin ligase. Mol Cell Biol 25: 1087510894.
  • Wang J, Hoshijima M, Lam J, Zhou Z, Jokiel A, Dalton ND, Hultenby K, Ruiz-Lozano P, Ross J Jr, Tryggvason K, Chien KR. 2006. Cardiomyopathy associated with microcirculation dysfunction in laminin alpha4 chain-deficient mice. J Biol Chem 281: 213220.
  • Wang X, Quail E, Hung N-J, Tan Y, Ye H, Costa RH. 2001. Increased levels of forkhead box m1b transcription factor in transgenic mouse hepatocytes prevents age-related proliferation defects in regenerating liver. Proc Natl Acad Sci USA 98: 1146811473.
  • Wang X, Kiyokawa H, Dennewitz MB, Costa RH. 2002. The Forkhead Box m1b transcription factor is essential for hepatocyte DNA replication and mitosis during mouse liver regeneration. Proc Natl Acad Sci USA 99: 1688116886.
  • Watt AJ, Battle MA, Li J, Duncan SA. 2004. GATA4 is essential for formation of the proepicardium and regulates cardiogenesis. Proc Natl Acad Sci USA 101: 1257312578.
  • Wells J, Farnham PJ. 2002. Characterizing transcription factor binding sites using formaldehyde crosslinking and immunoprecipitation. Methods 26: 4856.
  • Wells NJ, Watanabe N, Tokusumi T, Jiang W, Verdecia MA, Hunter T. 1999. The C-terminal domain of the Cdc2 inhibitory kinase Myt1 interacts with Cdc2 complexes and is required for inhibition of G(2)/M progression. J Cell Sci 112: 33613371.
  • Wilkins BJ, De Windt LJ, Bueno OF, Braz JC, Glascock BJ, Kimball TF, Molkentin JD. 2002. Targeted disruption of NFATc3, but not NFATc4, reveals an intrinsic defect in calcineurin-mediated cardiac hypertrophic growth. Mol Cell Biol 22: 76037613.
  • Ye H, Kelly TF, Samadani U, Lim L, Rubio S, Overdier DG, Roebuck KA, Costa RH. 1997. Hepatocyte nuclear factor 3/fork head homolog 11 is expressed in proliferating epithelial and mesenchymal cells of embryonic and adult tissues. Mol Cell Biol 17: 16261641.
  • Zaffran S, Frasch M. 2002. Early signals in cardiac development. Circ Res 91: 457469.
  • Zhao YY, Gao XP, Zhao YD, Mirza MK, Frey RS, Kalinichenko VV, Wang IC, Costa RH, Malik AB. 2006. Endothelial cell-restricted disruption of FoxM1 impairs endothelial repair following LPS-induced vascular injury. J Clin Invest 116: 23332343.