MAD1 and c-MYC regulate UBF and rDNA transcription during granulocyte differentiation
2004; Springer Nature; Volume: 23; Issue: 16 Linguagem: Inglês
10.1038/sj.emboj.7600335
ISSN1460-2075
AutoresGretchen Poortinga, Katherine M. Hannan, Hayley J. Snelling, Carl R. Walkley, Anna Jenkins, Kerith Sharkey, Meaghan Wall, Yves Brandenburger, Manuela Palatsides, Richard B. Pearson, Grant A. McArthur, Ross D. Hannan,
Tópico(s)Chronic Lymphocytic Leukemia Research
ResumoArticle29 July 2004free access MAD1 and c-MYC regulate UBF and rDNA transcription during granulocyte differentiation Gretchen Poortinga Gretchen Poortinga Division of Research, Peter MacCallum Cancer Centre, St Andrews Place, East Melbourne, Victoria, Australia Search for more papers by this author Katherine M Hannan Katherine M Hannan Division of Research, Peter MacCallum Cancer Centre, St Andrews Place, East Melbourne, Victoria, Australia Search for more papers by this author Hayley Snelling Hayley Snelling Division of Research, Peter MacCallum Cancer Centre, St Andrews Place, East Melbourne, Victoria, Australia Search for more papers by this author Carl R Walkley Carl R Walkley Division of Research, Peter MacCallum Cancer Centre, St Andrews Place, East Melbourne, Victoria, Australia Department of Medicine, St Vincent's Hospital, University of Melbourne, Fitzroy, Victoria, Australia Search for more papers by this author Anna Jenkins Anna Jenkins Division of Research, Peter MacCallum Cancer Centre, St Andrews Place, East Melbourne, Victoria, Australia Search for more papers by this author Kerith Sharkey Kerith Sharkey Division of Research, Peter MacCallum Cancer Centre, St Andrews Place, East Melbourne, Victoria, Australia Search for more papers by this author Meaghan Wall Meaghan Wall Division of Research, Peter MacCallum Cancer Centre, St Andrews Place, East Melbourne, Victoria, Australia Search for more papers by this author Yves Brandenburger Yves Brandenburger Fox Chase Cancer Center, Human Genetics Program, Philadelphia, PA, USA Search for more papers by this author Manuela Palatsides Manuela Palatsides Division of Research, Peter MacCallum Cancer Centre, St Andrews Place, East Melbourne, Victoria, Australia Search for more papers by this author Richard B Pearson Richard B Pearson Division of Research, Peter MacCallum Cancer Centre, St Andrews Place, East Melbourne, Victoria, Australia Department of Biochemistry and Molecular Biology, University of Melbourne, Parkville, Victoria, Australia Search for more papers by this author Grant A McArthur Corresponding Author Grant A McArthur Division of Research, Peter MacCallum Cancer Centre, St Andrews Place, East Melbourne, Victoria, Australia Department of Medicine, St Vincent's Hospital, University of Melbourne, Fitzroy, Victoria, Australia Division of Haematology/Medical Oncology, Peter MacCallum Cancer Centre, St Andrews Place, East Melbourne, Victoria, Australia Search for more papers by this author Ross D Hannan Corresponding Author Ross D Hannan Division of Research, Peter MacCallum Cancer Centre, St Andrews Place, East Melbourne, Victoria, Australia Department of Biochemistry and Molecular Biology, University of Melbourne, Parkville, Victoria, Australia Division of Haematology/Medical Oncology, Peter MacCallum Cancer Centre, St Andrews Place, East Melbourne, Victoria, Australia Search for more papers by this author Gretchen Poortinga Gretchen Poortinga Division of Research, Peter MacCallum Cancer Centre, St Andrews Place, East Melbourne, Victoria, Australia Search for more papers by this author Katherine M Hannan Katherine M Hannan Division of Research, Peter MacCallum Cancer Centre, St Andrews Place, East Melbourne, Victoria, Australia Search for more papers by this author Hayley Snelling Hayley Snelling Division of Research, Peter MacCallum Cancer Centre, St Andrews Place, East Melbourne, Victoria, Australia Search for more papers by this author Carl R Walkley Carl R Walkley Division of Research, Peter MacCallum Cancer Centre, St Andrews Place, East Melbourne, Victoria, Australia Department of Medicine, St Vincent's Hospital, University of Melbourne, Fitzroy, Victoria, Australia Search for more papers by this author Anna Jenkins Anna Jenkins Division of Research, Peter MacCallum Cancer Centre, St Andrews Place, East Melbourne, Victoria, Australia Search for more papers by this author Kerith Sharkey Kerith Sharkey Division of Research, Peter MacCallum Cancer Centre, St Andrews Place, East Melbourne, Victoria, Australia Search for more papers by this author Meaghan Wall Meaghan Wall Division of Research, Peter MacCallum Cancer Centre, St Andrews Place, East Melbourne, Victoria, Australia Search for more papers by this author Yves Brandenburger Yves Brandenburger Fox Chase Cancer Center, Human Genetics Program, Philadelphia, PA, USA Search for more papers by this author Manuela Palatsides Manuela Palatsides Division of Research, Peter MacCallum Cancer Centre, St Andrews Place, East Melbourne, Victoria, Australia Search for more papers by this author Richard B Pearson Richard B Pearson Division of Research, Peter MacCallum Cancer Centre, St Andrews Place, East Melbourne, Victoria, Australia Department of Biochemistry and Molecular Biology, University of Melbourne, Parkville, Victoria, Australia Search for more papers by this author Grant A McArthur Corresponding Author Grant A McArthur Division of Research, Peter MacCallum Cancer Centre, St Andrews Place, East Melbourne, Victoria, Australia Department of Medicine, St Vincent's Hospital, University of Melbourne, Fitzroy, Victoria, Australia Division of Haematology/Medical Oncology, Peter MacCallum Cancer Centre, St Andrews Place, East Melbourne, Victoria, Australia Search for more papers by this author Ross D Hannan Corresponding Author Ross D Hannan Division of Research, Peter MacCallum Cancer Centre, St Andrews Place, East Melbourne, Victoria, Australia Department of Biochemistry and Molecular Biology, University of Melbourne, Parkville, Victoria, Australia Division of Haematology/Medical Oncology, Peter MacCallum Cancer Centre, St Andrews Place, East Melbourne, Victoria, Australia Search for more papers by this author Author Information Gretchen Poortinga1, Katherine M Hannan1, Hayley Snelling1, Carl R Walkley1,2, Anna Jenkins1, Kerith Sharkey1, Meaghan Wall1, Yves Brandenburger3, Manuela Palatsides1, Richard B Pearson1,4, Grant A McArthur 1,2,5 and Ross D Hannan 1,4,5 1Division of Research, Peter MacCallum Cancer Centre, St Andrews Place, East Melbourne, Victoria, Australia 2Department of Medicine, St Vincent's Hospital, University of Melbourne, Fitzroy, Victoria, Australia 3Fox Chase Cancer Center, Human Genetics Program, Philadelphia, PA, USA 4Department of Biochemistry and Molecular Biology, University of Melbourne, Parkville, Victoria, Australia 5Division of Haematology/Medical Oncology, Peter MacCallum Cancer Centre, St Andrews Place, East Melbourne, Victoria, Australia ‡These two authors contributed equally to this work *Corresponding authors. Molecular Oncology Laboratory, Trescowthick Research Laboratories, Peter MacCallum Cancer Centre, St Andrew's Place, East Melbourne 3002, Victoria, Australia. Tel.: +61 3 9656 1195; Fax: +61 3 9656 1411; E-mail: [email protected] Control Laboratory, Trescowthick Research Laboratories, Peter Mac Callum Cancer Centre, St Andrew's Place, east Melbourne 3002, Victoria, Australia. Tel.: +61 3 9656 1747; Fax: +61 3 9656 1411; E-mail: [email protected] The EMBO Journal (2004)23:3325-3335https://doi.org/10.1038/sj.emboj.7600335 PDFDownload PDF of article text and main figures. ToolsAdd to favoritesDownload CitationsTrack CitationsPermissions ShareFacebookTwitterLinked InMendeleyWechatReddit Figures & Info The regulation of cell mass (cell growth) is often tightly coupled to the cell division cycle (cell proliferation). Ribosome biogenesis and the control of rDNA transcription through RNA polymerase I are known to be critical determinants of cell growth. Here we show that granulocytic cells deficient in the c-MYC antagonist MAD1 display increased cell volume, rDNA transcription and protein synthesis. MAD1 repressed and c-MYC activated rDNA transcription in nuclear run-on assays. Repression of rDNA transcription by MAD1 was associated with its ability to interact directly with the promoter of upstream binding factor (UBF), an rDNA regulatory factor. Conversely, c-MYC activated transcription from the UBF promoter. Using siRNA, UBF was shown to be required for c-MYC-induced rDNA transcription. These data demonstrate that MAD1 and c-MYC reciprocally regulate rDNA transcription, providing a mechanism for coordination of ribosome biogenesis and cell growth under conditions of sustained growth inhibition such as granulocyte differentiation. Introduction In mammals, cell growth is tightly coupled to the cell division cycle of rapidly proliferating cells (Pardee, 1989). However, in other contexts, such as cell hypertrophy or during the reduction in cell mass associated with differentiation, these processes become uncoupled. Cell growth is a complex process involving the synthesis of macromolecules such as rRNA and protein as well as the generation of energy through anabolic pathways. A central component of cell growth is protein synthesis, necessitating the generation of functional ribosomes (Peculis, 2002). The requirement for regulation of ribosome biogenesis is evident in studies showing that mutations in genes encoding key components of the ribosome-biosynthetic pathway result in both reduced cell growth and size (Jorgensen et al, 2002). Despite the biological importance of cell growth in normal development and in diseases such as cancer, the molecules that coordinate these processes in mammalian systems remain poorly defined. Ribosome biogenesis and subsequent increased protein synthesis in response to growth stimuli require the coordinated synthesis of numerous molecular components of the functional ribosome by all three RNA polymerases (reviewed in Larminie et al, 1998). RNA polymerase I transcribes multiple copies of the gene that encodes the 45S ribosomal RNA precursor of the 18S, 5.8S and 28S rRNA within the nucleoli (rDNA transcription). RNA polymerase III transcribes the 5S component of rRNA and various tRNA molecules that are required for translation. RNA polymerase II transcribes mRNAs that encode for a large number of ribosomal subunit proteins. Finally, post-translational mechanisms, such as phosphorylation of the ribosomal subunit protein S6 by S6 kinase, also regulate the function of the ribosome (Dufner and Thomas, 1999). Normal development requires mechanisms that acutely respond to the demand for growth as well as mechanisms for the sustained regulation of growth in processes such as differentiation. During the process of differentiation, many cell types undergo an arrest in the G1 phase of the cell cycle. Arrest of the cell division cycle is often accompanied by a reduction in cell growth and ribosome biogenesis. Indeed, some cell types such as granulocytes reduce their size during differentiation, suggesting that growth is being inhibited disproportionately to division during the final cell cycles prior to terminal differentiation. Elucidating the molecular mechanisms that control this tightly orchestrated process and how it is linked to ribosome biogenesis is an important question to be addressed. The MAX network of transcription factors is comprised of a group of bHLH-Zip proteins that form heterodimers with the bHLH-Zip protein MAX. These proteins include the MYC family of transcriptional activators (c, N and L-MYC) and the MAD family of transcriptional repressors. Both the MYC and MAD families of proteins have been implicated in the regulation of cell growth in addition to their established role in cell division. MYC in both Drosophila and mammalian cells promotes cell growth and protein synthesis (Iritani and Eisenman, 1999; Johnston et al, 1999; Schuhmacher et al, 1999). Consistent with this, characterization of transcriptional targets of c-MYC has revealed a number of genes involved in the promotion of various aspects of cell growth (Coller et al, 2000; Guo et al, 2000; Boon et al, 2001). In contrast, the MYC antagonist MAD1 reduces cell size when overexpressed and reduces expression of a number of molecules implicated in certain aspects of ribosome biogenesis and protein translation (Iritani et al, 2002). Recent studies have specifically implicated MYC in the regulation of ribosome biogenesis. First, MYC can regulate expression of ribosomal subunit proteins that are transcribed by RNA polymerase II (Guo et al, 2000; Kim et al, 2000; Shiio et al, 2002). Second, MYC can directly regulate gene transcription mediated by RNA polymerase III via interaction with TFIIIB (Gomez-Roman et al, 2003). Third, MYC regulates a series of proteins important in nucleolar function and rRNA processing that are transcribed by RNA polymerase II (Greasley et al, 2000; Zeller et al, 2001; Schlosser et al, 2003). While these mechanisms link proteins of the MAX network to ribosome biogenesis, we demonstrate here that MAD1 and c-MYC can regulate the final key component and major rate-limiting step of ribosome biogenesis: transcription of the rDNA genes. MAD1 and c-MYC were able to directly regulate the expression of upstream binding factor (UBF), an HMG-box protein whose expression facilitates rDNA transcription in vivo. siRNA was used to block UBF accumulation and this prevented c-MYC from fully activating rDNA transcription. These data indicate that members of the MAX network of transcription factors regulate rDNA transcription, providing a mechanism for the coordination of ribosome biogenesis and cell growth during granulocyte differentiation. Results MAD1 negatively regulates cell growth and rDNA transcription during granulocyte differentiation Granulocytic cells deficient in MAD1 undergo extra cell divisions before terminally differentiating (Foley et al, 1998); thus, we sought to determine if MAD1 also regulates cell growth during granulocyte differentiation. We examined whole bone marrow from Mad1-null and wild-type mice for differences in cell volume and observed an increased volume only in cells of the myeloid fraction from Mad1-null mice (Figure 1A). To determine the subpopulation of granulocytic cells contributing to this increase in cell volume, we isolated both the immature and mature granulocytic cell fractions and found that the immature granulocytes from Mad1-null animals displayed a statistically significant greater volume than wild-type cells (Figure 1B). We also looked at the total RNA content of the immature fraction, gating on G0/G1 cells to exclude any cell cycle differences, and found an increase in total RNA in cells from Mad1-null mice (Figure 1C). Over 80% of total cellular RNA is rRNA (Paule, 1998), suggesting that Mad1-null cells may have elevated levels of rRNA. As rRNA is transcribed within the nucleolus and the rate of rRNA synthesis is related to nucleolar size (Derenzini et al, 2000), we determined the size of nucleoli in granulocytic cells from Mad1-null and wild-type mice. Indeed, cells from Mad1-null mice displayed larger nucleoli (2.3-fold) compared to wild-type cells (Figure 1D). Figure 1.MAD1 regulates cell growth during granulocyte differentiation. (A) Representative cell volume profiles (left) and pooled data (right) of whole bone marrow cells from wild-type (WT; Mad1+/+) and Mad1-null (Mad1−/−) mice were determined using a Sysmex CDA500 system. Results are the mean±s.e.m. from four WT and seven Mad1−/− mice. ‡P=0.002 compared to cells from WT mice. (B) Immature granulocytes (CD11b/Gr-1dim staining) were purified from WT and Mad1−/− mouse bone marrow using FACS as described (Walkley et al, 2002). Representative cell volume profiles (left) and pooled data (right) of granulocytes (n=4 for each genotype) are shown. *P<0.05 compared to cells from WT mice. (C) Total RNA content of immature granulocytes in the G0/G1 phase of the cell cycle. Cell populations were purified as in (B), stained for RNA and DNA content using acridine orange and gated on G0/G1. Representative RNA staining profiles (left) and pooled data (right) of granulocytes (n=4 WT, 6 Mad1−/−) are shown. *P<0.05 compared to WT mice. (D) Size of granulocytic nucleoli from WT and Mad1−/− mice as demonstrated by Ag-NOR staining. Arrowheads point to nucleolar regions of granulocytes (left). Total area of Ag-NOR-stained nucleoli was calculated (right) (n=10 WT, 20 Mad1−/−). *P<0.05 compared to WT mice. Download figure Download PowerPoint To further investigate the possibility that loss of MAD1 leads to a higher rate of rRNA synthesis, we looked at a major rate-limiting step in rRNA accumulation: transcription of the 45S rRNA precursor by RNA polymerase I (Paule, 1998). To determine if rates of rDNA transcription were altered in granulocytes from Mad1-null mice, we used an ex vivo culture system where primary granulocytes in whole bone marrow were stimulated with stem cell factor (SCF) and G-CSF. Cells from Mad1-null mice had higher rates of synthesis of total RNA (34% increase) and rDNA transcription (2.5-fold) as measured by incorporation of [3H]uridine into RNA and nuclear run-on analyses of the 45S gene (loading of RNA polymerase I onto the 45S rDNA gene), respectively (Figure 2A and B). This was associated with an increase in cell volume (10% increase) and rate of total protein synthesis (2.4-fold) as determined by biosynthetic labeling with [35S]methionine (Figure 2C and D). Significantly, cell proliferation rates remained the same for both wild-type and Mad1-null cells (Figure 2E). To rule out differences in ploidy, DNA content was measured and found to be equivalent for wild-type and Mad1-null cells (data not shown). These data demonstrate that, under these culture conditions, the MAX network contributes to the regulation of growth and rDNA transcription in granulocytes independent of cell proliferation rate. Figure 2.MAD1 regulates RNA synthesis, rDNA transcription and protein synthesis in granulocytes. Cultures of primary granulocytes from wild-type (WT; Mad1+/+) and Mad1-null (Mad1−/−) mice were stimulated with SCF and G-CSF. Cells were analyzed after 5 days of culture and equal cell numbers were used for all assays. (A) RNA synthesis as measured by incorporation of [3H]uridine into total cellular RNA (n=4 for each genotype). *P<0.05 compared to WT mice. (B) rDNA transcription was measured by nuclear run-on analysis of the 45S rRNA precursor (n=4 WT, 4 Mad1−/−). *P<0.05 compared to WT mice. (C) Cell volume of cultured granulocytes (n=8 WT, 7 Mad1−/−). *P<0.05 compared to WT mice. (D) Protein synthesis as measured by incorporation of [35S]methionine into total cellular protein (n=4 WT, 5 Mad1−/−). *P<0.05 compared to WT mice. (E) Proliferation rate as determined by granulocyte concentration 5 days after wild-type and Mad1−/− cultures were seeded with 1 × 105 cells/10 ml of culture medium (n=7 WT, 7 Mad1−/−). Download figure Download PowerPoint MAD1 and c-MYC can regulate rDNA transcription in various cell types The above studies demonstrated that loss of MAD1 correlated with increased rDNA transcription in granulocytes. We therefore examined if enforced expression of MAD1 could regulate rDNA transcription as measured by nuclear run-on analyses of the 45S gene. The assays were performed in NIH3T3 cells: a well-characterized system for the study of rDNA transcription that allows for retroviral-mediated expression of potential regulatory proteins. Indeed, NIH3T3 cells infected with a retroviral vector expressing MAD1 displayed 40% lower transcription of the 45S gene compared to control cells (P<0.001) (Figure 3A). Figure 3.Regulation of rDNA transcription by MAD1 and c-MYC. (A) Proliferating cultures of NIH3T3 cells were infected with pBabe and pBabe-MAD1 expression vectors. Cells were then analyzed for endogenous rDNA transcription by nuclear run-on assay. Results are the mean±s.e.m. of four independent experiments. ***P<0.001 compared to pBabe cells. (B) Cultures of pBabe-MYC-ER NIH3T3 cells were incubated in DMEM containing 0.5% serum for 24 h (SS) and then stimulated with 10% serum (▪), 4-OHT (200 nM) () or vehicle (EtOH) () for 12 and 24 h before being analyzed for endogenous rDNA transcription by run-on assays (n=5; *P<0.05 and **P<0.01 compared to SS cells); or (C) 24 h for expression of the 45S rRNA precursor by Northern blot (β-actin control) and qRT-PCR (normalized to β-2-microglobulin expression). (D) Unstimulated (nonhypertrophic) cultures of primary neonatal cardiomyocytes were transfected with pSMECAT, pCMV-β-Gal (see Supplementary Figure 1) and an increasing amount of a c-MYC expression vector. After 24 h, cells were assayed for CAT activity (n=5). *P<0.05 and **P<0.01 compared to cells transfected with the empty expression vector (% control). (E) Cultures of primary neonatal cardiomyocytes were transfected with pSMECAT, pCMV-β-Gal (see Supplementary Figure 1) and an increasing amount of an MAD1 expression vector and then stimulated with the hypertrophic agent endothelin-1 (10−7 M). After 24 h, cells were assayed for CAT activity (n=5). *P<0.05 and **P<0.01 compared to hypertrophic cells transfected with the empty expression vector (% control). (F) Cultures of pBabe-MYC-ER MPRO granulocytic cells were induced to differentiate (for 4 days) and then stimulated with 4-OHT (200 nM) for 24 h before being assayed for 45S RNA expression by Northern blot (GAPDH control) and qRT-PCR (normalized to GAPDH expression). At 0 and 24 h, BrdU incorporation was assayed to determine the % of cells in S phase. Download figure Download PowerPoint While MAD:MAX complexes bind DNA and repress transcription, MYC:MAX complexes bind the same canonical sites and typically activate transcription (Eisenman, 2001). Given this, we sought to determine if c-MYC could activate rDNA transcription. NIH3T3 cells were infected with a retroviral vector expressing the c-MYC-ER fusion protein, an inducible form of c-MYC that can be rapidly activated in cells by the addition of 4-OH-tamoxifen (4-OHT) (Eilers et al, 1991; Littlewood et al, 1995). Cells expressing c-MYC-ER were cultured in media containing low serum, c-MYC was activated by the addition of 4-OHT and the rate of rDNA transcription was analyzed. Following induction of c-MYC activity, nuclear run-on analyses of the 45S gene demonstrated a significant increase (70%) in rDNA transcription as compared to control serum starved cells (P<0.01) that approximated the fold induction observed with serum (Figure 3B). Since the number of nuclei assayed per time point were normalized for DNA content before measurement of RNA polymerase I transcription, elevated gene copy number could not account for the increased rDNA transcription. In addition, Northern and quantitative real-time PCR (qRT-PCR) analyses of 45S rRNA precursor expression levels in these cells demonstrated a 2- to 3-fold increase in c-MYC-induced 45S expression when normalized to control transcripts (Figure 3C). To exclude the possibility that the ability of MAD1 to repress and c-MYC to activate rDNA transcription was an indirect consequence of the ability of these proteins to regulate cell proliferation, we examined the effect of expression of MAD1 or c-MYC on rDNA transcription in primary cultures of terminally differentiated neonatal cardiomyocytes. These cells were transfected with an rDNA reporter gene, pSMECAT (Hannan et al, 1996b), allowing us to examine nonproliferative growth (hypertrophic growth) that is dependent on increased rates of rDNA transcription (Brandenburger et al, 2001). c-MYC conferred a dose-dependent activation of rDNA transcription (up to 2.5-fold) in unstimulated cardiomyocytes (Figure 3D), consistent with studies demonstrating that conditional cardiomyocyte-specific overexpression of c-MYC stimulates hypertrophic growth (Xiao et al, 2001). Conversely, MAD1 significantly inhibited rDNA transcription to below basal levels in myocytes undergoing hypertrophy in response to the growth stimulant endothelin-1 (Luyken et al, 1996) (Figure 3E). Under the same conditions, c-MYC and MAD1 had no effect on a cotransfected control reporter gene, pCMV-β-Gal (Supplementary Figure 1). We also assayed the effect of enforced c-MYC expression on endogenous rDNA transcription rates in differentiated murine MPRO granulocytes, a model system for in vitro differentiation. MPRO cells infected with the c-MYC-ER-expressing retrovirus were induced to differentiate for 4 days, then c-MYC was activated by addition of 4-OHT for 24 h before expression of the 45S rRNA precursor was examined. Northern blot and qRT-PCR analyses demonstrated a 2.3-fold increase in expression of the 45S precursor, while cell cycle analysis showed that the cells remain arrested with approximately 3% of cells in S phase (Figure 3F). After c-MYC induction, cells also remained morphologically differentiated (data not shown). Together these findings indicate that the effects of c-MYC and MAD1 on rDNA transcription occur independently of cell division and cell type. The rDNA transcription factor UBF is regulated by c-MYC and MAD1 Transcription from rDNA genes is tightly regulated during changing cellular states such as the switch from quiescence to proliferation (Grummt et al, 1976; Paule, 1998), during cellular differentiation (Larson et al, 1993; Cavanaugh et al, 1995) or from basal growth to hypertrophy (Hannan et al, 1996a). The transcription factor UBF is an important regulator of rDNA transcription and, like proteins of the MAX network, expression of UBF is tightly regulated in many systems in response to changing cellular growth requirements (Larson et al, 1993; Cavanaugh et al, 1995). For example, differentiation of L6 myoblasts correlates with a simultaneous decrease in UBF expression and rDNA transcription (Larson et al, 1993). Conversely, serum refeeding of serum-deprived NIH3T3 fibroblasts leads to a rapid accumulation of UBF mRNA and protein (Supplementary Figure 2), which precedes the activation of rDNA transcription (Glibetic et al, 1995). We therefore examined expression of UBF during differentiation of the human HL-60 and murine MPRO granulocytic cell lines and correlated this with expression of c-MYC and MAD1. As previously described, expression of c-MYC is reduced and MAD1 is induced during granulocyte differentiation (Figure 4A). In parallel to the induction of MAD1, UBF protein (Figure 4A) and mRNA (Figure 4B) were significantly reduced during granulocyte differentiation. Strikingly, in HL-60 cells, many transcripts including β-actin (Figure 4B) reduced on differentiation. However, UBF mRNA is more tightly regulated, being absolved by day 1, consistent with the effects on protein levels. Figure 4.Regulation of c-MYC, MAD1 and UBF during granulocyte differentiation. (A) Human (HL-60) and mouse (MPRO) granulocytic cell lines were induced to differentiate with ATRA (for 1 and 2 days) and AGN194024 (for 2 and 4 days) respectively as compared to undifferentiated cells (day 0). Protein from cells was analyzed by Western blot for c-MYC, MAD1 and UBF expression. α-Tubulin was used as a protein loading control and ratios of UBF to α-tubulin were calculated. (B) Cells were differentiated as indicated in (A) and UBF mRNA levels were analyzed by Northern blot. Lanes are equally loaded with total RNA as demonstrated by β-actin expression and/or stained for 18S and 28S rRNA. d0=day 0, d1=day 1, d2=day 2 and d4=day 4. Download figure Download PowerPoint UBF regulates rDNA transcription and binds to multiple sites across the rDNA gene including the proximal promoter (Moss and Stefanovsky, 2002; O'Sullivan et al, 2002). We therefore performed chromatin immunoprecipitation (ChIP) assays on undifferentiated and differentiated MPRO cells to determine if binding of UBF to the rDNA gene was likewise reduced during granulocyte differentiation. Sites within three regions of the rDNA gene, all of which bind varying amounts of UBF, were examined: the 5′-enhancer sequence (ENH), the proximal rDNA promoter (upstream control element, UCE) and the externally transcribed spacer (ETS) (Figure 5A). As a negative control, we also examined UBF binding to the promoter of Lactoferrin, a gene that is transcriptionally activated in mature granulocytes. Binding of UBF to the rDNA gene was enriched, specifically at the ENH and UCE sites, in proliferating cells (day 0) whereas following differentiation (day 4) there was a marked reduction in UBF binding (Figure 5B and C) that correlated with a reduction in 45S precursor transcript (Figure 5D). Comparatively no UBF binding was demonstrated at the Lactoferrin promoter (Figure 5C). These results demonstrate that, as expression of c-MYC decreases and MAD1 increases, both the amount of UBF bound to the rDNA gene and the rate of rDNA transcription correlate with UBF expression during granulocyte differentiation. Figure 5.Regulation of UBF binding to the rDNA gene during granulocyte differentiation. MPRO cells were differentiated for 4 days and UBF binding to the rDNA gene was determined by ChIP assays using qRT-PCR. (A) Three regions of the murine rDNA gene were examined: the 5′ enhancer region (ENH); the upstream control element (UCE); and the 5′ external transcribed spacer (ETS). (B) Representative qRT-PCR amplification curves as displayed by ABI Prism 7000 are shown for the UCE amplicon. RS: control ChIP with preimmune rabbit sera. The arbitrary amplification threshold is depicted as the horizontal bar running across the graph. (C) Data presented as % of DNA site immunoprecipitated with anti-UBF compared to that present in total input DNA (based on 4% input; calculated as described in Frank et al, 2001). A site within the Lactoferrin promoter (lact2) was used as a negative control. Results are the mean±s.e.m. from three independent experiments. *P<0.05 compared to day-4 ChIPs. The representative ethidium bromide gel shows the amount of UCE product present after 19 cycles (within the exponential phase) of PCR of ChIP samples. (D) Cells were assayed for 45S rRNA expression by Northern blot
Referência(s)