Regulation of the cyclin D1 and cyclin A1 promoters by B-Myb is mediated by Sp1 binding sites

Gene 351 (2005) 171 – 180 www.elsevier.com/locate/gene

Regulation of the cyclin D1 and cyclin A1 promoters by B-Myb is mediated by Sp1 binding sites Thorsten Bartusel, Stephan Schubert, Karl-Heinz Klempnauer* Institut fu¨r Biochemie, Westfa¨lische-Wilhelms-Universita¨t Mu¨nster, Wilhelm-Klemm-Str. 2, D-48149 Mu¨nster, Germany Received 3 November 2004; received in revised form 21 February 2005; accepted 22 March 2005 Received by R. Di Lauro

Abstract B-Myb is a highly conserved member of the Myb family of transcription factors which plays an important role during the cell cycle. Previous work has shown that B-Myb is phosphorylated at several sites by cyclin A/Cdk2 in the early S-phase. These phosphorylations increase the transactivation potential of B-Myb by counteracting the repressive function of an inhibitory domain located at the carboxylterminus of B-Myb. As yet, only a few genes have been identified as B-Myb target genes. Previous work has suggested that the cyclin D1 gene might be regulated by B-Myb. Here, we have studied the effect of B-Myb on the promoter of the cyclin D1 gene. We show that B-Myb is a potent activator of the cyclin D1 promoter and that this activation is not mediated by Myb binding sites but rather by a group of Sp1 binding sites which have previously been shown to be crucial for cyclin D1 promoter activity. Our data show that the C-terminal domain of B-Myb is required for the activation of the cyclin D1 promoter and that this part of B-Myb interacts with Sp1. Finally, we have found that the promoter of the cyclin A1 gene is also activated by B-Myb by a Sp1 binding site-dependent mechanism. The effect of B-Myb on the promoters of the cyclin A1 and D1 genes is reminiscent of the mechanism that has been proposed for the autoregulation of the B-myb promoter by B-Myb, which also involves Sp1 binding sites. Taken together, our identification of two novel B-Myb responsive promoters whose activation by B-Myb does not involve Myb binding sites extends previous evidence for the existence of a distinct mechanism of transactivation by B-Myb which is dependent on Sp1 binding sites. The observation that this mechanism is not subject to the inhibitory effect of the C-terminal domain of B-Myb but rather requires this domain supports the notion that the Sp1 site-dependent mechanism is already active in the G1-phase prior to the phosphorylation of B-Myb by cyclin A/Cdk2. D 2005 Elsevier B.V. All rights reserved. Keywords: B-Myb; Sp1; Cyclin D1 promoter; Cyclin A1 promoter; Transactivation

1. Introduction The transcription factor B-Myb, a highly conserved member of the myb proto-oncogene family (Saville and Watson, 1998a; Joaquin and Watson, 2003), is expressed in a wide variety of dividing cells and tissues and appears to play a general role during cell proliferation. B-myb Abbreviations: GST, glutathione-S-transferase; SDS – PAGE, sodium dodecyl sulfate – polyacrylamide gel electrophoresis; tk, thymidine kinase; GFP, green fluorescent protein. * Corresponding author. Tel.: +49 251 8333203; fax: +49 251 8333206. E-mail address: [email protected] (K.-H. Klempnauer). 0378-1119/$ - see front matter D 2005 Elsevier B.V. All rights reserved. doi:10.1016/j.gene.2005.03.035

expression is downregulated in quiescent fibroblasts and subsequent re-entry into the cell cycle induces expression of B-myb in the late G1/early S-phase (Lam et al., 1992; Lam and Watson, 1993). During mouse embryogenesis B-myb expression is tightly linked to the proliferative activity of cells and tissues (Sitzmann et al., 1996). Downregulation of B-myb expression mediated by antisense oligonucleotides has been shown to inhibit cell proliferation (Arsura et al., 1992). The function of B-myb is regulated at multiple levels in a cell cycle dependent manner. B-myb mRNA expression reaches its maximum during the late G1 phase and is regulated by an E2F binding site within the B-myb promoter (Lam and Watson, 1993; Zwicker et al., 1996) as well as by

172

T. Bartusel et al. / Gene 351 (2005) 171 – 180

B-Myb itself (Sala et al., 1999). In addition to the transcriptional regulation of the B-myb gene, the activity of the B-Myb protein is controlled by post-translational modification and interactions with other proteins. Phosphorylation of B-Myb by cyclin A/Cdk2 in the early S-phase of the cell cycle stimulates the transactivation potential of BMyb by counteracting the repressive function of an inhibitory domain located at the carboxyl-terminus (Robinson et al., 1996; Lane et al., 1997; Sala et al., 1997; Ziebold et al., 1997; Saville and Watson, 1998b; Bartsch et al., 1999). Several B-Myb interacting proteins have been identified and shown to affect the B-Myb transactivation potential, including cyclin D1 (Horstmann et al., 2000; Schubert et al., 2004), poly-(ADP-ribose) polymerase (PARP) (Cervellera and Sala, 2000), p107 (Joaquin et al., 2002), nucleolin (Ying et al., 2000), p300 (Bessa et al., 2001; Johnson et al., 2002; Schubert et al., 2004), TAFII250 (Bartusel and Klempnauer, 2003) and N-CoR/SMRT (Masselink et al., 2001; Li and McDonnell, 2002). The interactions of some of these proteins, such as PARP (Santilli et al., 2001), N-CoR/SMRT (Li and McDonnell, 2002) and p300 (Schubert et al., 2004), appear to be modulated by the phosphorylation state of B-Myb suggesting that they are involved in the phosphorylation-dependent stimulation of B-Myb activity. In addition to its role as a cell-cycle regulated transcriptional activator B-Myb appears to exert some of its functions independently of its transcriptional activity. For example B-Myb has been shown to interact with cell cycle regulatory proteins, most notably p107, and to overcome a G1 cell cycle block imposed by p107 (Joaquin et al., 2002); this activity of B-Myb as well as its ability to overcome cell cycle arrest induced by p21(Waf1/Cip1) appears to be independent of the activation of specific target genes. B-Myb has been shown to either activate or repress the transcription of certain target genes in a cell-type and promoter-specific manner. For example, B-Myb activates the apolipoprotein J and Bcl-2 genes by acting through Myb-binding sites located in the promoters of these genes (Grassilli et al., 1999; Cervellera et al., 2000). In addition to this Myb-binding site-dependent activation B-Myb also employs other mechanisms to stimulate transcription. The activation of the HSP70 promoter by B-Myb appears to be mediated by the TATA-box and the heat-shock response element (Foos et al., 1993; Kamano and Klempnauer, 1997). The activation of the promoter of its own gene by B-Myb appears to be mediated by transcription factor Sp1 binding to sites located in the promoter region. Transcriptional repression by B-Myb has been observed for the type I collagen genes (Marhamati and Sonenshein, 1996; Kypreos et al., 1998; Cicchillitti et al., 2004). Different mechanisms have been proposed for these inhibitions, involving Myb binding sites as well as other proteins (Kypreos et al., 1999; Cicchillitti et al., 2004). In this work we have studied the effect of B-Myb on the promoters of the cyclin D1 and cyclin A1 genes. We became

interested in the cyclin D1 gene because previous work has suggested that the cyclin D1 gene might be regulated by BMyb (Sala and Calabretta, 1992). Our results show that BMyb activates the cyclin D1 promoter and that the activation is mediated by binding sites for transcription factor Sp1. Our data suggest a similar mechanism for the regulation of the cyclin A1 promoter. Taken together, our results highlight the role of Sp1 in mediating stimulatory effects of B-Myb on downstream targets.

2. Materials and methods 2.1. Eukaryotic expression vectors Expression vectors for full-length mouse B-Myb (pMuBMEx9A+) and for the deletion mutant B-Myb-D3 (pMuBMEx9a+D3) have been described before (Kamano et al., 1995). Expression vectors for HA-tagged full-length Sp1 and a Gal4-Sp1 fusion protein were obtained from H. Rotheneder and have been described before Haidweger et al., 2001). The p300 expression vector pCMV-p300CHA encodes full-length human p300 containing a C-terminal hemagglutinin tag and was a gift from R. Eckner (Eckner et al., 1994). 2.2. Reporter genes, transfections, luciferase and b-galactosidase assays The Myb-inducible reporter plasmid pGL3-3xATk-Luc has been described (Horstmann et al., 2000). Plasmids containing the chicken mim-1 (pGL-240Luc) and tom-1 promoters (pGL-144Luc) have been described (Ness et al., 1989; Burk et al., 1997). The human cyclin D1 promoter construct (pGL2-cycD1-luc) and the deletion series of the hamster cyclin D1 promoter were obtained form E.H. Wang and have been described (Dunphy et al., 2000; Hilton and Wang, 2003). p-186/-107 tk-81Luc contains sequences between 186 and 107 of the hamster cyclin D1 promoter fused upstream of the basal ( 81/+53) thymidine kinase promoter. The human cyclin A1 promoter constructs have been described (Mu¨ller et al., 1999). pCMVh was obtained from Clontech. QT6 cells were transfected as described (Burk et al., 1997). Cells were harvested 24 h after transfection. Preparation of cell extracts, luciferase and hgalactosidase assays was performed as described (Burk et al., 1997). 2.3. GST fusion proteins and in vitro binding assay GST-mouse B-Myb fusion proteins pGST/B-Myb-D1 (aminoacids 168-279), pGST/B-Myb-D2 (aminoacids 240376) and pGST/B-Myb-D3 (aminoacids 349-536) have been described (Schubert et al., 2004). pGST/B-Myb-DBD encodes a fusion protein of GST and the DNA-binding domain (aminoacids 1– 190) of B-Myb and was generated

T. Bartusel et al. / Gene 351 (2005) 171 – 180

30

luciferase activity

by cloning an approximately 600 bp EcoRI/SalI fragment from the 5¶ end of a mouse B-myb cDNA clone (pCDNA3MuBMEx9A+) between the EcoRI and XhoI sites of pGex4T-1. pGST/B-Myb-CT encodes a fusion protein of GST and the C-terminal domain (aminoacids 503– 704) of BMyb. This plasmid was generated by cloning an approximately 600 bp RsaI/XhoI fragment from the 3¶ end of a mouse B-myb cDNA clone between the SmaI and XhoI sites of pGex-5X-3. GST fusion proteins were prepared and used in pull-down experiments as described (Schubert et al., 2004).

173

20

10

2.4. In vivo co-immunoprecipitation To detect complexes of B-Myb and Sp1 in vivo, QT6 cells were transfected with expression vectors for B-Myb and HA-tagged Sp1. 24 h later the cells were lysed in ELB buffer (50 mM Tris – HCl, pH 7.5; 120 mM NaCl; 20 mM NaF; 1 mM Benzamidine; 1 mM EDTA; 6 mM EGTA; 15 mM sodium pyrophosphate; 1 mM PMSF; 0.1% NP-40) followed by centrifugation for 30 min at 14,000  g. Lysates were analyzed by SDS –PAGE or precipitated with antibodies against the HA-tag (HA.11, BabCO) followed by SDS –PAGE. Proteins were then detected by antibodies against B-Myb (N-19, St. Cruz Biotechnology).

B-Myb p300

- + - + - - ++

- + - + - - ++

pGL-240mim pGL-144luc

- + - + - - ++

- + - + - - ++

p3xAtk-Luc pGL-cycD1-Luc

Fig. 1. The stimulation of B-Myb transactivation by p300 is promoterdependent. QT6 cells were transfected with the indicated luciferase reporter genes (3 Ag each), the h-galactosidase plasmid pCMVh (0.2 Ag) and expression vectors for B-Myb (5 Ag) and p300 (7.5 Ag), as shown at the bottom. Luciferase and h-galactosidase activities were determined 24 h after transfection. Bars show the average activity of the luciferase reporter gene, normalized with respect to the h-galactosidase activity. The activity of the reporter genes in the absence of B-Myb was designated as 1. Thin lines show standard deviations.

3. Results 3.1. Differential effects of B-Myb and p300 on different Myb-inducible promoters We and others have shown before that B-Myb recruits the coactivator p300, resulting in a stimulation of the BMyb transactivation potential (Bessa et al., 2001; Johnson et al., 2002; Schubert et al., 2004). During these studies we analyzed the effect of B-Myb and p300 on the activity of a range of promoters and found that the ability of p300 to act as a coactivator of B-Myb was strongly dependent on the promoter context. Fig. 1 illustrates this observation. The activation by B-Myb of the promoters of the Myb target genes mim-1 (pGL-240mim) and tom-1 (pGL-144Luc) and of a construct containing the basal tk promoter fused to three copies of a Myb binding site (p3xAtk-Luc) was strongly stimulated by p300. The promoter of the cyclin D1 gene (pGL-cycD1-Luc) was also activated by B-Myb, however there was no additional stimulation of the promoter by p300. This observation suggested that B-Myb stimulates the cyclin D1 promoter by a different mechanism than that of the other promoters tested in this experiment. We became particularly interested in the promoter of the cyclin D1 gene because cyclin D1 itself has been shown to interact with B-Myb and to inhibit its transactivation potential (Horstmann et al., 2000; Schubert et al., 2004). The observation that B-Myb stimulates the cyclin D1 promoter therefore suggested the existence of a feed-back loop by which B-Myb increases the

expression of its own inhibitor. Furthermore, the activation of the cyclin D1 promoter was of interest because, unlike the other promoters tested, it does not have any obvious Myb binding sites. 3.2. Mapping of B-Myb responsive sites in the cyclin D1 promoter To understand how B-Myb activates the cyclin D1 promoter we mapped the part of the promoter which is responsive to B-Myb. We used a series of deletion mutants of the hamster cyclin D1 promoter kindly provided by E.H. Wang (Hilton and Wang, 2003) and co-transfected them with the B-Myb expression vector. Fig. 2 shows that deletion of the sequences between 186 and 107 bp upstream of the transcriptional start site reduced the B-Myb responsiveness of the promoter. This was observed in a series of promoter constructs that extend up to + 119 bp (left panel) as well as with shorter constructs extending up to position +33 bp relative to the major transcriptional start site (right panel). To determine whether this region affects the response to B-Myb of a heterologous promoter we fused it to the basal tk promoter and determined the effect of B-Myb on the activity of the resulting construct. As illustrated in Fig. 3, this part of the cyclin D1 promoter significantly increased the activity of the tk promoter in the absence of BMyb, presumably because several Sp1 binding sites are present on this part of the cyclin D1 promoter. Furthermore

174

T. Bartusel et al. / Gene 351 (2005) 171 – 180

1

2

3

4

5

6

25

3.3. Interaction of Sp1 with the C-terminal domain of B-Myb

20 luciferase activity

consistent with previous work which has shown that B-Myb needs to be activated by cyclin A/Cdk2 mediated phosphorylation or by C-terminal truncation in order to stimulate a promoter containing Myb binding sites. As expected, Cterminally truncated B-Myb stimulated this reporter gene better than full-length B-Myb. The cyclin D1/tk reporter gene showed a different response. On its own B-Myb already strongly increased the activity of the reporter gene; moreover, C-terminally truncated B-Myb was less active that full length B-Myb. These observations again illustrate that B-Myb stimulates the cyclin D1 promoter by a different mechanism than a promoter containing Myb binding sites.

The region between 186 and 107 bp of the cyclin D1 promoter does not contain Myb binding sites, instead there are three Sp1 binding sites which are conserved between the cyclin D1 promoters of different species. Sala

15

10

1

2

37.0

5

40

-186/+33

-109/+33

Fig. 2. Deletion analysis of the cyclin D1 promoter. The structure of the hamster cyclin D1 promoter is outlined schematically at the top. The positions of binding sites for several transcription factors are marked. QT6 cells were transfected with 3 Ag each of the indicated luciferase reporter genes, the h-galactosidase plasmid pCMVh (0.2 Ag) and expression vector for B-Myb (5 Ag, black columns). Control transfections (white columns) contained an equivalent amount of empty expression vector. Luciferase and h-galactosidase activities were determined 24 h after transfection. Bars show the average luciferase activity, normalized with respect to the h-galactosidase activity. Thin lines show standard deviations. The insert at the upper right side demonstrates that the amount B-Myb protein does not change with the reporter gene used. Aliquots of cells transfected as described above were analyzed by Western blotting using B-Myb antiserum. The following reporter genes were cotransfected with B-Myb: 715/+119 (lane 1); 443/ +119 (lane 2); 186/+119 (lane 3); 109/+119 (lane 4); 186/+33 (lane 5); 109/+33 (lane 6).

B-Myb activated the promoter construct containing the region between 186 and 107 bp of the cyclin D1 promoter to much higher levels than the original tk promoter. Thus, this experiment shows that this region of the cyclin D1 promoter increases B-Myb responsiveness when fused to another promoter. We also compared the activation by B-Myb of the recombinant cyclin D1/tk reporter gene and a tk promoter construct containing three copies of a Myb binding site (p3xAtk-Luc); this construct has been used in many of the previous studies to measure the transactivation potential of B-Myb. Fig. 3 shows that the 3xAtk-Luc reporter gene was not stimulated very efficiently by full-length B-Myb. This is

30 luciferase activity

-715/+119 -443/+119 -186/+119 -109/+119

19.2

20

8.6

10 3.5 1.2

1.7

2.8

ptk-81Luc

3.6

0.5

p-186/-107 tk-81Luc

p3xAtkLuc

Fig. 3. Sequences from 186 to 109 of the cyclin D1 promoter confer BMyb responsiveness to the basal tk promoter. QT6 cells were transfected with 3 Ag each of the luciferase reporter genes indicated at the bottom, the h-galactosidase plasmid pCMVh (0.2 Ag) and expression vectors for fulllength B-Myb (5 Ag, black columns) or C-terminally truncated B-Myb-D3 (8 Ag, hatched columns). Control transfections lacking B-Myb (white columns) contained 8 Ag of empty expression vector. Luciferase and hgalactosidase activities were determined 24 h after transfection. Bars show the average luciferase activity, normalized with respect to the hgalactosidase activity. Thin lines show standard deviations. For better comparison, the luciferase activities are also indicated by the numbers above each column. The insert at the top right of the figure shows a comparison of the amounts of full-length (lane 1) and C-terminally truncated B-Myb (lane 2). Aliquots of the cells transfected as described above were analyzed by SDS – PAGE and Western blotting using B-Myb antiserum.

T. Bartusel et al. / Gene 351 (2005) 171 – 180

A TAD

DBD

NRD

B-Myb-fl B-Myb-∆3

B B-Myb-fl

B-Myb-∆3

175

ground bands it appears that C-terminally truncated B-Myb (B-Myb-D3) co-precipitates only weakly or not at all with Sp1. To confirm the B-Myb/Sp1 interaction in vitro we performed GST pull-down assays using GST-fusion-proteins that contain different parts of B-Myb. Similar amounts of these proteins were incubated with extracts from cells transfected with an expression vector for a HA-tagged version of Sp1. Fig. 5 shows that the Sp1 protein was able to bind to the GST-protein containing the C-terminal domain of B-Myb. Weaker binding was also detected to domains D2 and D3 which comprise the central part of B-Myb. Thus, this experiment confirms that Sp1 interacts primarily with the C-terminal domain of B-Myb. However, the interaction

Sp1

A TAD

DBD

B-Myb-fl

1

GST-B-Myb-DBD

GST

GST-B-Myb-D1

GST

GST-B-Myb-D2

GST

GST-B-Myb-D3

GST

GST-B-Myb-CT

GST

190

168

B-Myb-∆3

NRD

B-Myb-fl

279

240

376

349

536

503

1 2

et al. (1999) have previously reported that B-Myb activates the promoter of its own gene via Sp1 binding sites and have suggested an interaction between B-Myb and Sp1 as the molecular basis of the activation of the B-myb promoter. The finding that the B-Myb responsive region of the cyclin D1 promoter contains several Sp1 binding sites suggested that B-Myb stimulates the cyclin D1 promoter also via Sp1. We were therefore interested to confirm the interaction between B-Myb and Sp1 and to map it with respect to the Cterminal part of B-Myb. Fig. 4 shows an in vivo coimmunoprecipitation analysis in which expression vectors for full-length or C-terminally deleted B-Myb were cotransfected with an Sp1 expression vector, followed by immunoprecipitation and Western blotting. This experiment clearly shows that full-length B-Myb can be co-precipitated together with Sp1. Although the bands corresponding to the partially deleted B-Myb mutants are obscured by back-

3 GS

T-B -

yb -D

My b-C T

2 yb -D

GS T-B -M

T-B -M

GS

-M yb -D

BD

1

Sp1

T-B

Fig. 4. Co-immunoprecipitation of B-Myb and Sp1. A. Schematic illustration of full-length or C-terminally truncated B-Myb. B. QT6 cells were transfected with expression vectors for full length or C-terminally truncated B-Myb, and HA-tagged Sp1, as indicated below. 24 h after transfection the cells were harvested and cell extracts were analyzed by Western blotting using antibodies against B-Myb (top). In addition, cell extracts were immunoprecipitated with antibodies against the HA-tag, followed by Western blotting of the immunoprecipitates using antibodies against the HA-tag (middle) or B-Myb (bottom). Relevant protein bands are marked.

B

GS

+ +

T

+ + -

-M yb -D

+ -

GS

+

GS T-B

B-Myb-fl

inp ut

+ B-Myb-∆3 Sp1

704

3 4 5 6

Fig. 5. Interaction of B-Myb and Sp1 in vitro. A. Schematic illustration of GST/B-Myb fusion proteins used. Numbers refer to aminoacids; DNAbinding (DBD), transactivation (TAD) and negative regulatory domains (NRD) are highlighted. B. Similar amounts of different GST fusion proteins were bound to glutathione-sepharose and incubated with extract from cells transfected with expression vector for HA-tagged Sp1. Bound proteins were analyzed by SDS – PAGE and Western blotting using HA-specific antibodies. Lane 1 shows a fraction of total cell extract (2.5%) used for each binding reaction. The bottom panel shows a Coomassie blue stained SDS – polyacrylamide gel of the different GST/B-Myb proteins. Lane 1: GST; lane 2: GST/B-Myb-DBD; lane 3: GST/B-Myb-D1; lane 4: GST/B-Myb-D2; lane 5: GST/B-Myb-D3; lane 6: GST/B-Myb-CT.

176

T. Bartusel et al. / Gene 351 (2005) 171 – 180

is quite weak raising the possibility that in vivo it is stabilized by other proteins. The results of the in vivo and in vitro binding experiments are, however, entirely consistent with our observation that the C-terminally truncated form of B-Myb (B-Myb-D3) stimulates the B-Myb responsive region of the cyclin D1 promoter much less than full-length B-Myb (see Fig. 3). To confirm the requirement of the Cterminal domain of B-Myb for cooperation with Sp1 we performed a mammalian two-hybrid experiment. In this experiment (Fig. 6) a Gal4-responsive reporter gene was stimulated by a Gal4-Sp1 construct and, additionally, different B-Myb constructs. As shown in Fig. 6, full-length B-Myb but not C-terminally truncated B-Myb increased the activity of the reporter gene significantly, consistent with the notion that the cooperation of B-Myb and Sp1 is dependent on the C-terminal domain of B-Myb. 3.4. B-Myb activates the human cyclin A1 promoter through Sp1 sites The cyclin A1 promoter has been shown to possess several Sp1 binding sites which are crucial for its activity

1

2

3

20 115 84

(Mu¨ller et al., 1999). We were therefore interested to investigate whether B-Myb also activates the cyclin A1 promoter via a Sp1 site-dependent mechanism. Cyclin A1 is specifically expressed during spermatogenesis and appears to play a role in meiosis (Yang et al., 1997), in contrast to cyclin A2 (also known as cyclin A) which is ubiquitously expressed in proliferating cells. Elevated levels of cyclin A1 have also been found in acute myeloid leukemia cell lines and in myeloid leukaemia samples from patients, suggesting that cyclin A1 contributes to leukemogenesis (Yang et al., 1999). We used reporter genes containing the human cyclin A1 and A2 promoters and cotransfected them with B-Myb expression vector. Whereas the cyclin A2 promoter, which lacks functional Sp1 binding sites, was not significantly stimulated by B-Myb (data not shown), the cyclin A1 promoter was activated by full-length B-Myb (Fig. 7). Interestingly, as in case of the cyclin D1 promoter the Cterminally truncated B-Myb stimulated the cyclin A1 promoter less well than full-length B-Myb, suggesting that the cyclin A1 and D1 promoters are activated by B-Myb by similar mechanisms. To address whether the stimulation of the cyclin A1 promoter is mediated by the Sp1 binding sites we used a set of reporter genes in which several or all of these sites had been mutated. Fig. 7 shows that the effect of B-Myb on the activity of the promoter responsiveness was strongly diminished when all of the Sp1 sites were mutated, demonstrating that the cyclin A1 promoter, like the cyclin D1 promoter, is activated by B-Myb through Sp1 sites.

luciferase activity

15

4. Discussion 10

5

Gal-Sp1 B-Myb B-Myb ∆3

-

+ -

+ -

+

+ + -

+ +

Fig. 6. Interaction of B-Myb and Sp1 in a two-hybrid experiment. QT6 cells were transfected with the Gal4-responsive luciferase plasmid pG5E4-38-luc (3 Ag), the h-galactosidase plasmid pCMVh (0.2 Ag) and expression vectors for Gal4-Sp1 (3 Ag) and full-length or C-terminally truncated BMyb protein constructs (2 and 6 Ag, respectively), as indicated at the bottom. Luciferase and h-galactosidase activities were determined 24 h after transfection. Bars show the average luciferase activity, normalized with respect to the h-galactosidase activity. Thin lines show standard deviations. The insert panel at the top shows a comparison of the amounts of the different B-Myb proteins. Aliquots of the cells transfected without BMyb expression vector (1), with 2 Ag of expression vector for full-length BMyb (2) or 6 Ag of expression vector for C-terminally truncated B-Myb (3) were analyzed by SDS – PAGE and Western blotting using B-Myb antiserum.

Sp1 is a member of a large family of transcription factors which have been implicated in a variety of biological functions (Philipsen and Suske, 1999; Suske, 1999; Kaczynski et al., 2003). It is widely held that Sp1 is involved in the basal transcription of many constitutive promoters, however there is also evidence suggesting that Sp1 family members are involved in growth control and cell cycle regulation (Black et al., 2001). Sp1 itself interacts with several proteins involved in cell cycle control, such as the transcription factor E2F (Karlseder et al., 1996), the retinoblastoma-related protein p107 (Datta et al., 1995), cyclin D1 (Adnane et al., 1999), p53 (Gualberto and Baldwin, 1995), Mdm2 (Johnson-Pais et al., 2001) and BMyb (Sala et al., 1999, and this work). In epithelial cells the expression of Sp1 reaches a maximum in the G1 phase of the cell cycle and a dominant-negative form of Sp1 causes a G1 cell cycle arrest (Grinstein et al., 2002). Furthermore, several reports have shown that the activity of Sp1 is modulated in a cell-cycle dependent fashion by cyclinA/ Cdk2 dependent phosphorylation (De Borja et al., 2001; Haidweger et al., 2001). Sp1 binding sites have been identified in the promoters of a number of genes regulated during the cell cycle. For example, the promoter of the murine thymidine kinase gene is stimulated by Sp1 together

T. Bartusel et al. / Gene 351 (2005) 171 – 180

177

luciferase activity 100

50

IV III

II

I

-1151

Luci

-190

Luci

-190mutI/II

Luci

-190mutIII/IV

Luci

1

-190mutI-IV

2

3

4

5

Luci

Fig. 7. Activation of the cyclin A1 promoter by B-Myb. The reporter genes used for transfection are shown schematically at the left. I to IV mark Sp1 binding sites. Mutant sites are marked by crosses. QT6 cells were transfected with the indicated luciferase reporter genes (3 Ag each), the h-galactosidase plasmid pCMVh (0.2 Ag) and expression vector for full-length B-Myb (5 Ag, black columns), C-terminally truncated B-Myb-D3 (10 Ag, hatched column) or empty expression vector (white columns). Luciferase and h-galactosidase activities were determined 24 h after transfection. Bars show the average luciferase activity, normalized with respect to the h-galactosidase activity. The luciferase activities of the wild-type cyclin A1 promoter constructs in the presence of B-Myb were designated as 100 in the upper and lower panels. Thin lines show standard deviations. The insert at the bottom right side demonstrates that the amount of BMyb protein does not change with the reporter gene used. Aliquots of cells transfected as described above were analyzed by Western blotting using B-Myb antiserum. The following reporter genes were cotransfected with full-length B-Myb: 1151 (lane 1); 190 (lane 2); 190mutI/II (lane 3); 190mutIII/IV (lane 4); 190mutI – IV (lane 5).

with E2F (Rotheneder et al., 1999). The relevance of Sp1 for cyclin D1 expression is well established. The Sp1 binding sites in the cyclin D1 promoter are highly conserved (Eto, 2000) and have been shown to be required for the transcriptional activation of the cyclin D1 gene following mitogenic stimulation (Nagata et al., 2001). Furthermore, expression of a dominant-negative Sp1 (Grinstein et al., 2002) or inhibition of Sp1 activity by Imperatorin, a coumarin derivative, blocks endogenous cyclin D1 expression (Sancho et al., 2004). In this work we have shown that B-Myb activates the cyclin D1 promoter and that this activation is not mediated by Myb binding sites but rather by the cluster of Sp1 binding sites in the cyclin D1 promoter. We have observed a similar requirement of Sp1 binding sites for B-Myb stimulation in case of the cyclin A1 promoter. Our data show that the C-terminal domain of B-Myb is required for the activation of both promoters and that this part of B-Myb, either directly or indirectly, interacts with Sp1. While these observations suggest that B-Myb and Sp1 cooperate they do not exclude the possibility that B-Myb also cooperates with other Sp1 family members, which may bind to the Sp1 binding sites. The mechanism uncovered by this work is highly reminiscent of the mechanism that has been proposed for the autoregulation of the B-myb promoter by B-Myb, which also involves Sp1 sites (Sala et al., 1999). Thus, our

identification of two novel B-Myb responsive promoters whose activation by B-Myb is dependent on Sp1 binding sites confirms and substantiates the notion that B-Myb affects target promoters by this novel mechanism. Previous work has shown that B-Myb overexpression leads to increased expression of the endogenous cyclin D1 mRNA (Sala and Calabretta, 1992); together with the data presented here this suggests that the cyclin D1 gene is a bona fide BMyb target regulated by B-Myb via Sp1 binding sites. Effects of B-Myb on the endogenous cyclin A1 gene have not yet been reported, however both genes are co-expressed during spermatogenesis. Unlike the cyclin A2 gene which is expressed in all proliferating cells the cyclin A1 gene is specifically expressed during spermatogenesis (Yang et al., 1997). In situ staining of seminiferous tubules for B-Myb (Sitzmann et al., 1996) and cyclin A1 (Mu¨ller-Tidow et al., 2001) shows very similar patterns, indicating that both genes are expressed during the same stages of spermatogenesis. Furthermore, Mu¨ller-Tidow et al. (2001) have demonstrated that GFP-positive cells from the testes of transgenic mice carrying a cyclin A1 promoter-driven GFP transgene express high levels of B-Myb whereas GFPnegative cells from the testes of these animals do not express B-Myb. These findings show that B-Myb and cyclin A1 are coexpressed during spermatogenesis. B-Myb may also be involved in cyclin A1 expression that has been

178

T. Bartusel et al. / Gene 351 (2005) 171 – 180

detected in acute myeloid leukemia cell lines and in myeloid leukaemia samples from patients (Yang et al., 1999). It appears that there are at least two sides of B-Mybs function as a transcriptional activator. On one hand, a large body of evidence has shown that B-Myb is a cell-cycle regulated transcription factor whose transactivation potential is affected positively by cyclin A/Cdk2 dependent phosphorylation (Robinson et al., 1996; Lane et al., 1997; Sala et al., 1997; Ziebold et al., 1997; Saville and Watson, 1998b; Bartsch et al., 1999) and negatively by Cdk-independent interaction with cyclin D1. The phosphorylation alleviates the inhibitory effect mediated by the C-terminal domain of B-Myb, thus permitting the activation of target promoters that possess Myb binding sites. Transactivation is greatly stimulated by interaction of B-Myb with the coactivator p300/CBP (Bessa et al., 2001; Johnson et al., 2002; Schubert et al., 2004), in addition it has been suggested that phosphorylation of B-Myb disrupts the binding of the corepressor N-CoR with the C-terminal domain of B-Myb (Li and McDonnell, 2002). Thus, the C-terminal domain of B-Myb appears to acts as a cell cycle sensor and directs BMyb activity to the late G1/early S-phase of the cell cycle. On the other hand, the novel Sp1 site-dependent activation mechanism clearly differs from the mechanism described above in that it requires Sp1 binding sites instead of Myb binding sites and in that is not subject to the inhibitory effect of the C-terminal domain of B-Myb, but actually requires this domain. This suggest that the Sp1 site-dependent mechanism may already be active at a time during the cell cycle prior to the phosphorylation of the C-terminal domain of B-Myb by cyclin A/Cdk2. Probably, this mechanism supports the transcriptional activation of the B-Myb gene during the G1-phase by creating an autoregulatory loop at the B-myb promoter. As pointed out above, cyclin D1 acts as a repressor of B-Myb activity which keeps B-Myb inactive as a binding site-dependent transactivator until the onset of the S-phase. We therefore suspect that the stimulation of the cyclin D1 gene by B-Myb is part of a feedback system by which the amounts of B-Myb and cyclin D1 are adjusted to each other during the G1 phase to ensure that sufficient cyclin D1 is available to titrate the increasing amount of BMyb and to prevent it from prematurely activating promoters containing Myb binding sites. Recent work has suggested a role for cyclin A1 in the response to DNA damage and it appears that the cyclin A1 gene is activated in many cell types following UV- or girradiation. This activation is mediated by SP1 sites present in the cyclin A1 promoter (Mu¨ller-Tidow et al., 2004). Interestingly, based on work in Drosophila melanogaster there is also emerging evidence for a possible role of B-Myb in the response of cells to DNA damage. Drosophila has a single myb gene which has been implicated in the maintenance of genomic stability (Katzen et al., 1998; Fitzpatrick et al., 2002; Manak et al., 2002) and recent work has shown that B-myb is able to complement a defective Drosophila myb gene (Davidson et al., 2005). It is therefore

possible that the activation of the cyclin A1 promoter by BMyb which we have described here is part of such a damage-response mechanism. In any case, our work clearly shows that B-Myb is able to transactivate target promoters via Sp1 sites and that this mechanism is not restricted to the B-myb gene itself but may play a more general role.

Acknowledgments We thank E. Wang, C. Mu¨ller-Tidow and H. Rotheneder for providing plasmids and A. Brehmer-Fastnacht and B. Michaelis for excellent technical assistance. This study was supported by a grant from the DFG (KL 461/9-2). References Adnane, J., Shao, Z., Robbins, P.D., 1999. Cyclin D1 associates with the TBP-associated factor TAF(II)250 to regulate Sp1-mediated transcription. Oncogene 18, 239 – 247. Arsura, M., Introna, M., Passerini, F., Mantovani, A., Golay, J., 1992. Bmyb antisense oligonucleotides inhibit proliferation of human hematopoietic cell lines. Blood 79, 2708 – 2716. Bartsch, O., Horstmann, S., Toprak, K., Klempnauer, K.-H., Ferrari, S., 1999. Identification of cyclin A/Cdk2 phosphorylation sites in B-Myb. Eur. J. Biochem. 260, 384 – 391. Bartusel, T., Klempnauer, K.-H., 2003. Transactivation mediated by B-Myb is dependent on TAF(II)250. Oncogene 22, 2932 – 2941. Bessa, M., Saville, M.K., Watson, R.J., 2001. Inhibition of cyclin A/Cdk2 phosphorylation impairs B-Myb transactivation function without affecting interactions with DNA or the CBP coactivator. Oncogene 20, 3376 – 3386. Black, A.R., Black, J.D., Azizkhan-Clifford, J., 2001. Sp1 and Kru¨ppel-like factor family of transcription factors in cell growth regulation and cancer. J. Cell. Physiol. 188, 143 – 160. Burk, O., Worpenberg, S., Haenig, B., Klempnauer, K.-H., 1997. tom-1, a novel v-Myb target gene expressed in AMV-and E26-transformed myelomonocytic cells. EMBO J. 16, 1371 – 1380. Cervellera, M.N., Sala, A., 2000. Poly(ADP-ribose) polymerase is a BMYB coactivator. J. Biol. Chem. 275, 10692 – 10696. Cervellera, M., et al., 2000. Direct transactivation of the anti-apoptotic gene apolipoprotein J (clusterin) by B-Myb. J. Biol. Chem. 275, 21055 – 21060. Cicchillitti, L., Jiminez, S.A., Sala, A., Saitta, B., 2004. B-Myb acts as a repressor of human Col1A1 collagen gene expression by interacting with Sp1 and CBF factors in scleroderma fibroblasts. Biochem. J. 378, 609 – 616. Datta, P.K., Raychaudhuri, P., Bagchi, S., 1995. Association of p107 with Sp1: genetically separable regions of p107 are involved in regulation of E2F- and Sp1-dependent transcription. Mol. Cell. Biol. 15, 5444 – 5452. Davidson, C.J., Tirouvanziam, R., Herzenberg, L.A., Lipsick, J.S., 2005. Functional evolution of the vertebrate Myb gene family: B-Myb, but neither A-Myb nor c-Myb, complements Drosophila Myb in hemocytes. Genetics 169, 215 – 229. De Borja, P.F., Collind, N.K., Du, P., Azizkhan-Clifford, J., Mudryj, M., 2001. Cyclin A-CDK phosphorylates Sp1 and enhances Sp1-mediated transcription. EMBO J. 20, 5737 – 5747. Dunphy, E.L., Johnson, T., Auerbach, S.S., Wang, E.H., 2000. Requirement for TAF(II)250 acetyltransferase activity in cell cycle progression. Mol. Cell. Biol. 20, 1134 – 1139. Eckner, R., et al., 1994. Molecular cloning and functional analysis of the adenovirus E1A-associated 300-kD protein (p300) reveals a

T. Bartusel et al. / Gene 351 (2005) 171 – 180 protein with properties of a transcriptional adaptor. Genes Dev. 8, 869 – 884. Eto, I., 2000. Molecular cloning and sequence analysis of the promoter region of mouse cyclin D1 gene: implication in phorbol ester-induced tumour promotion. Cell Prolif. 33, 167 – 187. Fitzpatrick, C.A, Sharkov, N.V., Ramsay, G., Katzen, A.L., 2002. Drosophila myb exerts opposing effectson S phase, promoting proliferation and suppressing endoreduplication. Development 129, 4497 – 4507. Foos, G., Natour, S., Klempnauer, K.-H., 1993. TATA-box dependent transactivation of the human HSP70 promoter by Myb protein. Oncogene 8, 1775 – 1782. Grassilli, E., Salomoni, P., Perrotti, D., Franseschi, C., Calabretta, B., 1999. Resistance to apoptosis in CTLL-2 cells overexpressing B-Myb is associated with B-Myb-dependent bcl-2 induction. Cancer Res. 59, 2451 – 2456. Grinstein, E., Jundt, F., Weinert, I., Wernet, P., Royer, H.-D., 2002. Sp1 as G1 cell cyclwe phase specific transcription factor in epithelial cells. Oncogene 21, 1485 – 1492. Gualberto, A., Baldwin Jr., A.S., 1995. p53 and Sp1 interact and cooperate in the tumor necrosis factor-induced transcriptional activation of the HIV-1 long terminal repeat. J. Biol. Chem. 270, 19680 – 19683. Haidweger, E., Novy, M., Rotheneder, H., 2001. Modulation of Sp1 activity by a cyclin A/CDK complex. J. Mol. Biol. 306, 201 – 212. Hilton, T.L., Wang, E.H., 2003. Transcription factor IID recruitment and Sp1 activation. Dual function of TAF1 in cyclin D1 transcription. J. Biol. Chem. 278, 12992 – 13002. Horstmann, S., Ferrari, S., Klempnauer, K.-H., 2000. Regulation of B-Myb activity by cyclin D1. Oncogene 19, 298 – 306. Joaquin, M., Watson, R.J., 2003. Cell cycle regulation by the B-Myb transcription factor. Cell. Mol. Life Sci. 60, 2389 – 2401. Joaquin, M., Bessa, M., Saville, M.K., Watson, R.J., 2002. B-Myb overcomes a p107-mediated cell proliferation block by interacting with an N-terminal domain of p107. Oncogene 21, 7921 – 7932. Johnson, L.R., et al., 2002. Effects of B-Myb on gene transcription: phosphorylation-dependent activity and acetylation by p300. J. Biol. Chem. 277, 4088 – 4097. Johnson-Pais, T., Degnin, C., Thayer, M.J., 2001. pRB induces Sp1 activity by relieving inhibition mediated by MDM2. Proc. Natl. Acad. Sci. U. S. A. 98, 2211 – 2216. Kaczynski, J., Cook, T., Urrutia, R., 2003. Sp1- and Kruppel-like transcription factors. Genome Biol. 4, 206. Kamano, H., Klempnauer, K.-H., 1997. B-Myb and cyclin D1 mediate heat shock element dependent activation of the human HSP70 promoter. Oncogene 14, 1223 – 1229. Kamano, H., Burk, B., Noben-Trauth, K., Klempnauer, K.-H., 1995. Differential splicing of the mouse B-myb gene. Oncogene 11, 2575 – 2582. Karlseder, J., Rotheneder, H., Wintersberger, E., 1996. Interaction of Sp1 with the growth- and cell cycle-regulated transcription factor E2F. Mol. Cell. Biol. 16, 1659 – 1667. Katzen, A.L., Jackson, J., Harmon, B.P., Fung, S.M., Ramsay, G., Bishop, J.M., 1998. Drosophila myb is required for the G2/M transition and maintenance of diploidy. Genes Dev. 12, 831 – 843. Kypreos, K.E., Nugent, M.A., Sonenshein, G.E., 1998. Basic fibroblast growth factor-induced decrease in type I collagen gene transcription is mediated by B-Myb. Cell Growth Differ. 9, 723 – 730. Kypreos, K.E., Marhamati, D.J., Sonenshein, G.E., 1999. B-Myb represses trans-activation of the Col5A2 collagen promoter indirectly via inhibition of binding of factors interacting with positive elements within the first exon. Matrix Biol. 18, 275 – 285. Lam, E.W., Watson, R.J., 1993. An E2F-binding site mediates cell-cycle regulated repression of mouse B-myb transcription. EMBO J. 12, 2705 – 2713. Lam, E.W., Robinson, C., Watson, R.J., 1992. Characterization and cell cycle-regulated expression of mouse B-myb. Oncogene 7, 1885 – 1890.

179

Lane, S., Farlie, P., Watson, R., 1997. B-Myb function can be markedly enhanced by cyclin A-dependent kinase and protein truncation. Oncogene 14, 2445 – 2453. Li, X., McDonnell, D.P., 2002. The transcription factor B-Myb is maintained in an inhibited state in target cells through its interaction with the nuclear corepressors N-CoR and SMRT. Mol. Cell. Biol. 11, 3663 – 3673. Manak, J.R., Mitiku, N., Lipsick, J.S., 2002. Mutation of the Drosophila homologue of the Myb protooncogene causes genomic instability. Proc. Natl. Acad. Sci. U. S. A. 99, 7438 – 7443. Marhamati, D.J., Sonenshein, G.E., 1996. B-Myb expression in vascular smooth muscle cells occurs in a cell cycle-dependent fashion and downregulates promoter activity of type I collagen genes. J. Biol. Chem. 271, 3359 – 3365. Masselink, H., Vastenhouw, N., Bernards, R., 2001. B-myb rescues rasinduced premature senescence, which requires its transactivation domain. Cancer Lett. 171, 87 – 101. Mu¨ller, C., Yang, R., Beck-von-Peccoz, L., Idos, G., Verbeek, W., Koeffler, H.P., 1999. Cloning of the cyclin A1 genomic structure and characterization of the promoter region. GC boxes are essential for cell cycleregulated transcription of the cyclin A1 gene. J. Biol. Chem. 274, 11220 – 11228. Mu¨ller-Tidow, C., et al., 2001. Cyclin A1 directly interacts with B-myb and cyclin A1/cdk2 phosphorylate B-myb at functionally important serine and threonine residues: tissue-specific regulation of B-myb function. Blood 97, 2091 – 2097. Mu¨ller-Tidow, C., et al., 2004. The cyclin A1-CDK2 complex regulates DNA double-strand break repair. Mol. Cell. Biol. 24, 8917 – 8928. Nagata, D., et al., 2001. Transcriptional activation of the cyclin D1 gene is mediated by multiple cis-elements, including SP1 sites and a cAMPresponsive element in vascular endothelial cells. J. Biol. Chem. 276, 662 – 669. Ness, S.A., Marknell, A., Graf, T., 1989. The v-myb oncogene product binds to and activates the promyelocyte-specific mim-1 gene. Cell 59, 1115 – 1125. Philipsen, S., Suske, G., 1999. A tale of three fingers: the family of mammalian Sp/XKLF transcription factors. Nucleic Acids Res. 27, 2991 – 3000. Robinson, C., Light, Y., Groves, R., Mann, D., Marais, R., Watson, R., 1996. Cell-cycle regulation of B-Myb protein expression: specific phosphorylation during the S phase of the cell cycle. Oncogene 12, 1855 – 1864. Rotheneder, H., Geymayer, S., Haidweger, E., 1999. Transcription factors of the Sp1 family: interaction with E2F and regulation of the murine thymidine kinase promoter. J. Mol. Biol. 293, 1005 – 1015. Sala, A., Calabretta, B., 1992. Regulation of BALB/c 3T3 fibroblast proliferation by B-myb is accompanied by selective activation of cdc2 and cyclin D1 expression. Proc. Natl. Acad. Sci. U. S. A. 89, 10415 – 10419. Sala, A., et al., 1997. Activation of human B-MYB by cyclins. Proc. Natl. Acad. Sci. U. S. A. 94, 532 – 536. Sala, A., et al., 1999. B-MYB transactivates its own promoter through SP1binding sites. Oncogene 18, 1333 – 1339. Sancho, R., et al., 2004. Imperatorin inhibits HIV-1 replication through an Sp1-dependent pathway. J. Biol. Chem. 279, 37349 – 37359. Santilli, G., Cervellera, M.N., Johnson, T.K., Lewis, R.E., Iacobelli, S., Sala, A., 2001. PARP co-activates B-MYB through enhanced phosphorylation at cyclin/cdk2 sites. Oncogene 20, 8167 – 8174. Saville, M.K., Watson, R.J., 1998a. B-Myb: a key regulator of the cell cycle. Adv. Cancer Res. 72, 109 – 140. Saville, M.K., Watson, R.J., 1998b. The cell-cycle regulated transcription factor B-Myb is phosphorylated by cyclin A/Cdk2 at sites that enhance its transactivation properties. Oncogene 17, 2679 – 2689. Schubert, S., Horstmann, S., Bartusel, T., Klempnauer, K.-H., 2004. The co-operation of B-Myb with the coactivator p300 is orchestrated by cyclins A and D1. Oncogene 23, 1392 – 1404.

180

T. Bartusel et al. / Gene 351 (2005) 171 – 180

Sitzmann, J., Noben-Trauth, K., Kamano, H., Klempnauer, K.-H., 1996. Expression of B-Myb during mouse embryogenesis. Oncogene 12, 1889 – 1894. Suske, G., 1999. The Sp-family of transcription factors. Gene 238, 291 – 300. Yang, R., Morosetti, R., Koeffler, H.P., 1997. Characterization of a second human cyclin A that is highly expressed in testis and in several leukemic cell lines. Cancer Res. 58, 985 – 990. Yang, R., et al., 1999. Cyclin A1 expression in leukemia and normal hematopoietic cells. Blood 93, 2067 – 2074.

Ying, G.G., Proost, P., van Damme, J., Bruschi, M., Introna, M., Golay, J., 2000. Nucleolin, a novel partner for the Myb transcription factor family that regulates their activity. J. Biol. Chem. 275, 4152 – 4158. Ziebold, U., Bartsch, O., Marais, R., Ferrari, S., Klempnauer, K.-H., 1997. Phosphorylation and activation of B-Myb by cyclin A-Cdk2. Curr. Biol. 7, 253 – 260. Zwicker, J., Liu, N., Engeland, K., Lucibello, F.C., Muller, R., 1996. Cell cycle regulation of E2F site occupation in vivo. Science 271, 1595 – 1597.