(Rb) Protein Upregulates Expression of the Ifi202 Gene Encoding

Home , JUNB, JunD

Oncogene (2003) 22, 4775–4785 & 2003 Nature Publishing Group All rights reserved 0950-9232/03 $25.00 www.nature.com/onc ORIGINAL PAPERS Retinoblastoma (Rb) protein upregulates expression of the Iﬁ202 gene encoding an interferon-inducible negative regulator of cell growth

Hong Xin1, Rocky Pramanik1 and Divaker Choubey*,1

1Department of Radiation Oncology, Stritch School of Medicine, Loyola University Medical Center, 2160 South First Avenue, Building No. 1, Mail code: 114B, Maywood, IL 60153, USA

Studies have indicated that ectopic expression of p202, an p202 increase in mouse embryonic fibroblast (MEF) cell interferon (IFN)-inducible retinoblastoma (Rb)-binding lines under reduced serum conditions and addition of protein, in cultured cells retards cell proliferation and growth factors, such as basic fibroblast growth factor modulates cell survival. Consistent with a role of p202 in (bFGF) or platelet-derived growth factor (PDGF), to cell cycle regulation, levels of p202 increase in cells culture medium inhibits the increase in p202 levels arrested in the G0/G1 phase of cell cycle after withdrawal (Geng et al., 2000). Additionally, levels of p202 also of serum growth factors. However, a role for p202 in cell increase during differentiation (induced by reduced growth arrest remains to be defined. Moreover, it remains serum conditions) of C2C12 murine myoblasts in vitro unclear how levels of p202 are upregulated during the cell (Datta et al., 1998). Interestingly, accumulation of growth arrest. Here, we report that Rb upregulates functional p53 in cells represses transcription of Ifi202 expression of Ifi202 gene. We found that basal as well gene (D’Souza et al., 2001). Remarkably, decreases in as IFN-induced levels of p202 were significantly higher in p202 levels increase susceptibility to apoptosis induced wild-type (Rb þ / þ ) mouse embryonic fibroblasts (MEFs) by different stimuli (Choubey, 2000; Choubey and than isogenic RbÀ/À MEFs. Consistent with the regulation Kotzin, 2002). Together, these observations support of Ifi202 gene by Rb, expression of functional Rb, but not the idea that cell growth-regulatory pathways, which a pocket mutant of it, stimulated the activity of a reporter regulate cell cycle progression and cell survival, can also whose expression was driven by the 50-regulatory region of regulate p202 levels. However, molecular mechanisms Ifi202 gene. Importantly, the stimulation by Rb was by which levels of p202 are upregulated remain to be dependent, in part, on a JunD/AP-1 DNA-binding site investigated. More importantly, it remains unclear how present in the 50-regulatory region of the Ifi202 gene. p202 contributes to cell growth arrest and increased cell Moreover, basal levels of p202 were significantly higher in survival. wild-type (JunD þ / þ ) than isogenic JunDÀ/À MEFs. Constitutive as well as inducible overexpression Additionally, we found that increased expression of p202 (about twofold above the basal levels) of p202 in a potentiated the Rb-mediated inhibition of cell growth and variety of cultured cells results in retardation of cell mutations in the Rb-binding motif (LxCxE) of p202 proliferation (Choubey, 2000). Additionally, the induci- significantly reduced cell survival. Together, our observa- ble expression of p202 in MEFs results in increases in tions support the idea that the transcriptional activation of p21WAF1 levels, a potent inhibitor of cyclin-dependent Ifi202 gene by Rb/JunD may be important for the kinases (Cdks; Weinberg, 1995), and reduced phosphor- regulation of cell growth and survival. ylation of Rb (Gutterman and Choubey, 1999). Oncogene (2003) 22, 4775–4785. doi:10.1038/sj.onc.1206780 Furthermore, the amino-acid sequence of p202 contains the Rb binding motif LxCxE to bind to Rb pocket (and Keywords: interferon-inducible p202; cell cycle; cell other ‘pocket’ proteins) (Choubey et al., 1989; Choubey survival; Rb; JunD/AP-1 and Lengyel, 1995), raising the possibility that p202- mediated inhibition of cell growth, in part, depends on its ability to bind to Rb pocket. Surprisingly, viral oncoproteins, such as adenovirus-encoded E1A protein and SV40 large T antigen, which are known to target Rb Introduction for functional inactivation (Weinberg, 1995), also target p202 for functional inactivation (Xin et al., 2001), Expression of p202, an interferon (IFN) and differentia- raising the possibility that p202 could also inhibit cell tion-inducible phosphoprotein (52 kDa) from the p200- growth, in part, by mechanisms independent of Rb. protein family, is detectable in a variety of mouse tissues The 50-regulatory region of Ifi202 gene has been well and cultured cells (Johnstone and Trapani, 1999; characterized (Samanta et al., 1986; Gribaudo et al., Choubey, 2000; Choubey and Kotzin, 2002). Levels of 1987). Like other housekeeping genes, the Ifi202 promoter sequence lacks the TATA and CAAT boxes. *Correspondence: D Choubey; E-mail: [email protected] The promoter contains potential DNA-binding sites for Received 19 December 2002; revised 7May 2003; accepted 12 May 2003 several transcription factors, including the p53 (D’Souza Rb upregulates p202 expression HXinet al 4776 et al., 2001) and AP-1 (Geng et al., 2000; Xin et al., 2000). Together, these observations suggest that JunD 2003). Our previous analyses indicated that Ifi202 gene acts as a negative regulator of cell growth and increases promoter contains potential AP-1 DNA-binding sites cell survival. However, the molecular mechanisms by (Geng et al., 2000). Interestingly, mutations of one of which JunD inhibits cell growth and increases cell the AP-1 DNA-binding site (the 202AP-1CS1) in Ifi202- survival remain presently unclear. promoter markedly reduced the transcription of a Rb binds to the Jun family of proteins, including reporter gene (Xin et al., 2003). Moreover, gel-shift JunD, and activates transcription through AP-1 sites analyses revealed that an oligonucleotide containing the (Nead et al., 1998; Nishitani et al., 1999). As Rb plays an 202AP-1CS1 sequence could bind to AP-1 transcription important role in G0/G1 cell growth arrest (Weinberg, factor complex containing either JunD (Geng et al., 1995) and increases in levels of JunD under reduced 2000) or c-Jun (Xin et al., 2003). These observations serum conditions correlate well with increases in p202 support the idea that AP-1 transcription factor contain- levels (Geng et al., 2000), we investigated whether Rb ing either JunD or c-Jun in the complex can regulate could activate transcription of Ifi202 gene through AP-1 transcription of Ifi202-gene through the 202AP-1CS1 sites. Here, we report that Rb upregulates expression of site. Ifi202 gene, in part, through JunD/AP-1. Importantly, Rb regulates progression from G0/G1 to S phase of we demonstrate that increased levels of p202 potentiate the cell cycle (Weinberg, 1995; Zheng and Lee, 2001) the Rb-mediated inhibition of cell growth. Moreover, and plays a role in the differentiation of muscle cells p202 cooperates with Rb to increase cell survival. Our (Yee et al., 1998; Lipinski and Jacks, 1999). Regulation observations support the idea that transcriptional of the G0/G1 phase of the cell cycle entails the binding of activation of Ifi202 gene by Rb/JunD is a part of Rb to a number of transcription factors, such as the E2F positive feedback loop, which contributes to cell growth family (Nevins, 1998), ATF-2 (Li and Wicks, 2001), c- arrest and cell survival. Myc (Rustgi et al., 1991). The most studied activity is the ability of Rb to bind to E2F (Nevins, 1998; Harbour and Dean, 2000). Rb binds the transactivation domain Results of E2F through the A/B pocket and inhibits the transcriptional activity of E2Fs. Phosphorylation of Rb upregulates expression of p202 Rb by cyclin-dependent kinases during the latter part of G1 phase of cell cycle results in the release of ‘free’ E2F To test whether Rb regulates expression of p202, we and the upregulation of E2F-responsive genes, many of compared basal, as well as induced levels of p202 which are required for DNA synthesis (Weinberg, 1995; (induced by IFN or serum starvation), in wild-type Harbour and Dean, 2000). Binding of viral oncopro- (Rb þ / þ ) and isogenic RbÀ/À MEFs. As shown in teins, such as SV40 large T antigen or adenovirus E1A, Figure 1, basal, as well as IFN-induced levels of p202 to Rb releases ‘free’ E2F (Nevins, 1998). were significantly higher in wild-type than Rb-null Additionally, a direct or indirect interaction of MEFs (compare lane 1 with lane 4 and lane 2 with lane hypophosphorylated Rb with transcription factors C/ 5, respectively). As reported earlier (Almasan et al., EBPa, b, and g (Chen et al., 1996a; Charles et al., 2001), 1995), serum starvation of RbÀ/À MEFs resulted in NF-IL6 (Chen et al., 1996b), AP-2 (Decary et al., 2002), extensive cell death as compared to Rb þ / þ MEFs. AP-1 (Nead et al., 1998; Nishitani et al., 1999), ATF-2 Therefore, we could not compare levels of p202 between (Li and Wicks, 2001), Sp-1 or Sp-3 (Udvadia et al., Rb þ / þ and RbÀ/À MEFs under reduced serum condi- 1993) has been demonstrated to augment their tran- tions. Furthermore, a comparison of the steady-state scriptional activity and promote tissue-specific gene expression. Rb-inducible genes include TGF-b1 (Kim et al., 1991) and p21WAF1 (Decesse et al., 2001). The AP-1 family of transcription factors is ubiqui- tously expressed and is involved in a variety of cell responses, including cell cycle arrest (Karin, 1995; Karin et al., 1997). There are a number of genes regulated by AP-1 factors and their control is complex since members of the Jun family (c-Jun, JunB and JunD) can homodimerize or heterodimerize with Fos and ATF family members. Different Jun-Jun or Jun-Fos com- Figure 1 Rb regulates expression of p202. Left panel. Levels of plexes bind to the recognition site 50-TGAC/GTCA-30 p202 were analysed by immunoblotting in extracts derived from wild-type (lanes 1–3) and isogenic Rb-null MEFs (lanes 4 and 5) with different affinities (Karin, 1995). JunD differs without any treatment (lanes 1 and 4), after IFN treatment markedly from the c-Jun and JunB, both of which are (1000 U/ml, 24 h; lanes 2 and 5) or after serum starvation (0.5% induced upon mitogenic stimulation of cells (Castellazzi serum, 36 h; lane 3). Levels of p68 protein, a crossreacting protein et al., 1991). Overexpression of JunD in immortalized detected by antiserum to p202 (Choubey and Lengyel, 1993) served fibroblasts causes retardation of cell proliferation (Pfarr as a control for protein loading amounts. Right panel. Levels of 202 mRNA were analysed by semiquantitative RT–PCR in total et al., 1994). Moreover, JunD protects cells from RNA isolated from wild-type (lane 1) and isogenic Rb-null (lane 2) apoptotic responses to stress stimuli, by acting as a MEFs. Levels of actin mRNA served as control for equal amounts modulator of p53-dependent apoptosis (Weitzman et al., of starting RNA in two samples

Oncogene Rb upregulates p202 expression HXinet al 4777 levels of the 202 RNA levels between wild-type and Rb- after serum stimulation (Lembo et al., 1995), suggesting null MEFs by a semiquantitative RT–PCR revealed that that functional inactivation of Rb by growth factors, the steady-state levels of the 202 RNA are two- to resulting in a decrease in p202 levels may be needed for threefold higher in wild-type than Rb-null MEFs transition of cells from the G0/G1 phase of cell cycle to (Figure 1). These observations suggest that Rb could the S phase. Therefore, upregulation of p202 by upregulate expression of p202 by activating transcrip- functional Rb in above experiments prompted us to tion of Ifi202 gene. test whether levels of p202 decrease during a transition Levels of p202 increase in mouse fibroblasts, which from G0/G1 phase (quiescence) to S phase of cell cycle. harbor functional Rb, under reduced serum conditions As shown in Figure 3a, levels of the 202 RNA were and addition of serum growth factors abrogates the significantly higher in growth arrested (arrested in the increase (Geng et al., 2000). Therefore, we tested G0/G1 phase) AKR 2B cells (compare 202 RNA levels whether the functional inactivation of Rb by viral between lanes 2 and 1). However, serum-stimulation of oncoproteins (it does not involve inactivation of Rb by cells for about 13 h significantly reduced 202 RNA levels phosphorylation), such as SV40 large T antigen or (compare lane 3 with lane 2). Similarly, a semiquanti- adenovirus-encoded E1A protein, affects p202 levels. As tative RT–PCR (specific to 202 mRNA) revealed that shown in Figure 2, expression of SV40 large T antigen in 202 RNA levels decreased after serum-stimulation of NIH 3T3 cells (these cells express functional Rb; cells for about 13 h (compare lane 3 with lane 1 in Choubey and Lengyel, 1995) resulted in reduced basal Figure 3b). More importantly, the decrease in 202 levels of p202 (compare lane 4 with lane 1). Similarly, mRNA levels coincided with the progression of cells expression of E1A in NIH 3T3 cells also resulted in from G0/G1 phase into the S phase of cell cycle, as significant reduction in levels of p202 (compare lane 7 indicated by a significant decrease in the G1/S ratio from with lane 1). Importantly, induction of p202 by serum 15.6 in growth arrested cells (lane 1) to 3.6 in serum- starvation of cells also decreased significantly after stimulated cells (see Figure 3b, lane 3). A decrease in expression of SV40 large T (compare lane 6 with lane 3) or E1A (compare lane 9 with lane 3). Similarly, IFN treatment of cells expressing SV40 large T antigen (compare lane 5 with lane 2) or E1A (compare lane 8 with lane 2) also did not result in appreciable increases in levels of p202. Together, these observations provide support for the idea that the functional Rb upregulates p202 levels and growth promoting signals, such as serum growth factors or viral oncoproteins, which functionally inactivate Rb, abrogate this increase in p202 levels.

Exit of cells from quiescence is associated with decreases in p202 levels Ectopic expression of p202 in NIH 3T3 cells signiﬁ- cantly delays exit of cells from G0/G1 phase (quiescence)

Figure 3 Levels of 202 mRNA and protein decrease during the transition of cells from quiescence to S phase of cell cycle. (a) Levels of 202 mRNA were analysed by Northern blotting in total RNA isolated from AKR-2B cells incubated in 10% serum (lane 1), 0.5% serum for 36 h (lane 2) or first incubated in 0.5% serum for 36 h and then serum-stimulated by 10% serum for 13 h (lane 3). The bottom panel shows the amount of RNA in an agarose gel used for Northern blotting. (b) Levels of 202 RNA were analysed by semiquantitative RT–PCR in total RNA isolated from AKR-2B cells incubated in 0.5% serum for 36 h (lane 1) or first incubated in Figure 2 Expression of SV40 large T antigen or adenovirus- 0.5% serum for 36 h and then serum-stimulated by 10% serum for encoded E1A protein in NIH 3T3 cells decreases basal as well as 8 h (lane 2) or 13 h (lane 3). Levels of actin mRNA served as control induced levels of p202. Levels of p202 were analysed by for equal amounts of starting RNA in samples. The numbers above immunoblotting in extracts derived from NIH 3T3 cells (lanes 1– indicate the ratio between the percentage of cells in the G1 phase of 3), NIH 3T3(T) cells expressing SV40 large T antigen (lanes 4–6) or cell cycle and the S phase of cell cycle. (c) Levels of p202, cyclin D1, NIH 3T3(E) cells expressing E1A (13S) protein without any E2F1 and actin were analysed by immunoblotting in extracts treatment (lanes 1, 4 and 7), after IFN treatment (1000 U/ml, 24 h; prepared from AKR-2B cells incubated in 10% serum (lane 1), lanes 2, 5 and 8) or after serum starvation (0.5% serum, 36 h; lanes 0.5% serum for 36 h (lane 2) or first incubated in 0.5% serum for 3, 6 and 9). Levels of actin protein served as a control for protein 36 h and then serum-stimulated by 10% serum for 8 h (lane 3) or loading amounts 13 h (lane 4)

Oncogene Rb upregulates p202 expression HXinet al 4778 levels of p202 was also evident after stimulation of growth-arrested cells with serum (Figure 3c, compare lane 4 with lane 2). Importantly, a decrease in p202 levels was accompanied by increases in levels of cyclin D1 and transcription factor E2F1, known indicators of progression of cells from G1 to S phase of cell cycle (Harbour and Dean, 2000). Similar results were obtained using NIH 3T3 cells (data not shown). These observations, along with other observations described above, are consistent with the possibility that Rb- induced increased expression of p202 under reduced serum conditions contributes to cell growth arrest and a decrease in p202 levels is necessary for timely exit of cells from the G0/G1 phase of cell cycle and progression of cells to the S phase.

Rb activates transcription of the Ifi202 gene To investigate how Rb transcriptionally activates expression of the Ifi202-gene, we performed promoter– reporter assays. As shown in Figure 4a, expression of a constitutively active form of Rb (mutant pD34; defective in phosphorylation by Cdks, therefore, constitutively ‘active’; see Hamel et al., 1992) in Rb-null MEFs stimulated the activity of the 202-luc-reporter in a dose- dependent manner. Furthermore, coexpression of the adenovirus-encoded E1A protein (13S), which binds to Rb in the pocket and inactivates its growth suppressor activity (Nevins, 1998), abrogated the Rb-induced stimulation of the activity of 202-luc-reporter (Figure 4b, compare column 3 with column 2). Similar results were obtained in isogenic wild-type cells (data not shown). Of note, expression of the pocket protein p107in Rb-null MEFs did not result in stimulation of the 202-luc-reporter activity (data not shown), raising the possibility that the pocket proteins differentially regulate expression of p202. As the pocket region of Rb is important for its Figure 4 Rb activates transcription of Ifi202 gene. (a)In promoter–reporter assays, Rb-null MEFs were transfected with growth-inhibitory function (Weinberg, 1995; Chen and increasing amounts of a plasmid allowing expression of Rb (mutant Wang, 2000; Pennaneach et al., 2001), we tested whether pD34). After 40–44 h of transfections, luciferase activity was expression of mutants of Rb (PM1 and PM2; defective determined. Bars indicate the standard deviation. (b) Cells were in the pocket domain function) also stimulates the transfected with a plasmid allowing expression of Rb (column 2) or activity of 202-luc-reporter. As shown in Figure 4c, plasmids allowing coexpression of Rb and adenovirus E1A (column 3). After 44 h of transfections, luciferase activity was expression of wild-type Rb stimulated the activity of the determined as described in the Materials and methods. Bars reporter, albeit less than the constitutively active Rb indicate the standard deviation. (c). The activity of 202-luc reporter (pD34; compare column 2 with column 3) used in after transfections of equal amounts of plasmid encoding wild-type experiments described above. Importantly, expression of Rb (column 2), a mutant (pD34) constitutively active form of Rb (column 3), a totally defective pocket mutant of Rb (N757F; a point mutant of Rb [PM1; Rb (N757F)], which is column 4) or a partially defective pocket mutant of Rb (L769D; totally defective in the LxCxE-binding pocket of column 5). Bars indicate the standard deviation Rb and lacks the ability to inhibit cell growth (Chen and Wang, 2000), did not result in stimulation of the activity of reporter (compare column 4 with Regulators of Rb also regulate expression of p202 column 1). As expected, expression of a mutant of Rb [PM2; Rb (L769D)], which is shown to be partially Transcriptional regulation of the Ifi202 gene by func- defective in the pocket functions (Chen and Wang, tional Rb, but not a pocket mutant of Rb, in above 2000), resulted in a moderate stimulation of the activity experiments raised the possibility that the physiological of 202-luc-reporter (compare column 5 with column 1). regulators of Rb, such as cyclin D1 and p16INK4a Together, these observations support the idea that Rb (Weinberg, 1995), could also regulate expression of transcriptionally activates expression of Ifi202 gene p202. As shown in Figure 5, expression of cyclin D1 in and the activation depends on the functional pocket mouse AKR 2B fibroblasts (these cells harbor a of Rb. functional Rb) (Choubey and Lengyel, 1995), in two

Oncogene Rb upregulates p202 expression HXinet al 4779 can bind to one of the potential AP-1 DNA-binding sites (the 202AP-1CS1 site) present in the Ifi202-gene in gel mobility shift assays (Geng et al., 2000); and (4) Rb binds to members of AP-1 family (c-Jun, JunD, and JunB) to activate transcription of a number of genes (Nead et al., 1998; Nishitani et al., 1999). Based on the above observations, we first tested whether expression of JunD could regulate basal levels and/or IFN-induced levels of p202. As shown in Figure 6, basal levels of p202 were significantly higher in wild-type (JunD þ / þ ) than JunD-null (JunDÀ/À) MEFs (compare lane 1 with lane 3). This observation is consistent with our earlier observation in which constitutive overexpression of mouse JunD in AKR- 2B fibroblast cell line resulted in upregulation of p202 levels (Geng et al., 2000). Moreover, IFN-induced levels Figure 5 Regulators of Rb, such as cyclin D1 and p16INK4a, also regulate expression of Ifi202 gene. Mouse AKR-2B cells were were comparable between wild-type and JunD-null cells transfected with increasing amounts (0.5 or 1.0 mg) of indicated (Figure 6, compare lane 2 with lane 4). These observa- expression plasmids as described in the Materials and methods. tions indicated that JunD positively regulates expression The total amount of plasmid in the assay was kept constant of Ifi202 gene. (1.0 mg). Bars indicate the standard deviation We next tested whether JunD could cooperate with Rb to activate transcription of Ifi202-gene. As shown in Figure 7a, expression of JunD alone stimulated the independent experiments, significantly (45–50%) re- activity of 202-luc-reporter about two-fold (compare pressed the activity of the 202-luc in a dose-dependent column 2 with column 1) in Rb-null MEFs. Similarly, manner (compare column 3 with column 1). However, expression of Rb (mutant pD34) alone also stimulated as expected, expression of p16INK4a resulted in a the activity about twofold (compare column 3 with significant increase (two- to threefold) in the activity of column 1). However, in several experiments coexpres- the 202-luc (compare column 3 with column 1). sion of JunD and Rb markedly (6.5-fold) increased the Similarly, expression of cyclin D1 in wild-type MEFs, activity of the 202-luc-reporter (compare column 4 with but not in Rb-null MEFs, resulted in 60–70% repression column 1), suggesting cooperation between JunD and of the 202-luc-reporter activity (data not shown). Rb to activate the transcription of Ifi202 gene. Together, these observations provide support for the The 50-regulatory region of Ifi202 gene contains four idea that the signaling pathways, which regulate the DNA binding sites for AP-1 (see Figure 7b). However, phosphorylation of Rb, also regulate expression of the the site 202-AP1CS1 was found to bind an AP-1 Ifi202-gene. complex with relatively higher affinity (Xin et al., 2003). Therefore, we mutated this AP-1 DNA-binding Rb and JunD/AP-1 cooperate to activate transcription of site by site-directed mutagenesis. As shown in Figure 7b, mutation in this AP-1 DNA binding site (the 202AP- Ifi202-gene 1CS1 site) in the Ifi202-gene promoter significantly To elucidate the molecular mechanisms, by which Rb (450%) decreased the activity of 202-luc reporter after transcriptionally activates expression of the Ifi202-gene, expression of JunD (compare the two columns in 2). we analysed the 50-regulatory region of Ifi202-gene for However, expression of Rb resulted in only moderate the presence of cis-elements through which Rb is known decreases in the activity of the reporter (compare the to transcriptionally activate gene expression (Zheng and Lee, 2001). The analyses revealed that the 50-regulatory region of Ifi202-gene contains potential DNA-binding sites for several transcription factors known to cooperate with Rb. These cis-elements also included potential DNA binding sites for transcription factor AP-1 (Geng et al., 2000; Xin et al., 2003). In order to identify cis-element(s) contributing to Rb-mediated transcriptional activation of the Ifi202-gene, we con- sidered the following observations: (1) under reduced serum conditions, an increase in the levels of p202 correlates with an increase in JunD levels (Geng et al., 2000); (2) ectopic expression of JunD in AKR-2B fibroblast cell line results in upregulation of p202 (Geng Figure 6 JunD regulates expression of p202. Levels of p202, JunD and actin were analysed by immunoblotting in extracts derived et al., 2000) and the activity of 202-luc-reporter is from wild-type (lanes 1 and 2) and isogenic JunD-null MEFs (lanes þ / þ significantly higher in MEFs derived from JunD 3 and 4) without any treatment (lanes 1 and 3) or after IFN than JunDÀ/À embryo (data not shown); (3) JunD/AP-1 treatment (1000 U/ml, 24 h; lanes 2 and 4)

Oncogene Rb upregulates p202 expression HXinet al 4780 p202 potentiates Rb-mediated inhibition of cell growth Binding of p202 to Rb, in part, depends on LxCxE Rb pocket binding motif (Choubey and Lengyel, 1995; Choubey, 2000). Moreover, overexpression of p202 in a variety of cells results in retardation of cell growth (Choubey, 2000; Choubey and Kotzin, 2002). Therefore, we tested whether p202 could potentiate Rb-mediated inhibition of cell proliferation. For this purpose, we chose human Saos-2 cells because these cells have a deletion in Rb gene (Huang et al., 1988) and have been used extensively to study cell growth inhibition by Rb in colony formation assays (Zheng and Lee, 2001). As shown in Figure 8, transfection of Saos-2 cells with a plasmid encoding p202, in two experiments, did not result in an inhibition of colony formation (as compared to vector-transfected cells). However, transfection of a plasmid-encoding mouse Rb (mRb) resulted in a signiﬁcant decrease (66%) in drug-resistant colony formation. Interestingly, cotransfection of plasmids encoding mRb and p202 resulted in a marked decrease (84%) in colony formation. Moreover, cotransfection of plasmids encoding mRb and a mutant of p202 (p202LxGxK mutant; LxCxE Rb-binding motif in p202 changed to LxGxK) did not result in potentiation of Rb- mediated inhibition of colony formation (data not shown). Together, these observations are consistent with the idea that the LxCxE motif-dependent binding of p202 to mRb pocket potentiates the Rb-mediated inhibition of colony formation in these cells.

Figure 7 Rb cooperates with JunD/AP-1 to activate transcription of Iﬁ202 gene. (a) In promoter–reporter assays, Rb-null MEFs were transfected with a plasmid-encoding mouse JunD (2 mg, column 1), Rb (2 mg, column 2) or JunD and Rb (2 mg each, respectively, column 4). The total amount of plasmid was kept constant to 4 mg with pCMV vector. After 44 h of transfections, luciferase activity was determined. Bars indicate the standard deviation. (b) Upper panel. Schematic presentation of the 50- regulatory region of the Iﬁ202 gene linked to the luciferase reporter gene. Potential AP-1 DNA binding sites in the 202-promoter are indicated as CS1, CS2, CS3 and CS4. The CS1 site (the 202AP- 1CS1 site), which was mutated, is indicated in black. Bottom panel. Rb-null MEFs were transfected with the indicated 202-luc-reporter plasmid along with a plasmid allowing expression of JunD (2 mg, column 2), Rb (5 mg, column 3) or both JunD and Rb (2 mg and 5 mg, respectively, column 4). Total amount of plasmid was kept constant to 7 mg in each assay. After 40–44 h of transfections, luciferase activity was determined as described in the Materials and methods. Bars indicate the standard deviation

two columns in 3), indicating that in the absence of the 202AP-1CS1 site Rb could stimulate transcription of Ifi202 gene through other Rb-responsive sites (see Discussion). Importantly, in several experiments mutation in the AP-1CS1 site resulted in a significant (50%) Figure 8 p202 potentiates Rb-mediated inhibition of colony decrease in synergistic activation of the activity of 202- formation. Human Saos-2 cells were transfected with equal luc-reporter after coexpression of JunD and Rb amounts of pCMV, pCMV-202, pCMV-mRb or both pCMV-202 (compare the two columns in 4). These experiments, and pCMV-mRb plasmids. Transfected cells were selected in G418 thus, indicated that Rb activates transcription of Ifi202- for about two weeks and the resistant colonies in plates were B gene in cooperation with the transcription factor JunD/ stained. The number of colonies ( 2 mm size) obtained in pCMV vector transfected plates is taken as 100%. The numbers indicate AP-1, in part, through the 202AP-1CS1 site in the Ifi202 the percent colonies obtained as compared to vector transfected gene. plates. A representative experiment is shown

Oncogene Rb upregulates p202 expression HXinet al 4781 p202 cooperates with Rb to increase cell survival column 1). Remarkably, expression of antisense to 202 mRNA in MCF-7cells sensitized these cells to UV-C- Studies indicate that expression of functional Rb induced apoptosis (compare column 3 with column 1). increases survival of cells in a variety of cell systems Next, we tested whether Rb-binding LxCxE motif in (Haput et al., 1995; Wang, 1997; Tan and Wang, 1998). p202 contributes to increased survival of cells after UV- Similarly, increased expression of p202 in a variety of C exposure. As shown in Figure 9b, expression of wild- cell systems correlates with resistance to apoptosis type p202 signiﬁcantly (60%) reduced cell death in (Choubey and Kotzin, 2002). Therefore, we tested MCF-7cells (compare column 2 with column 1). whether p202 could cooperate with Rb (and other However, expression of a mutant of p202, in which pocket proteins) to increase cell survival. For this Rb-binding motif LxCxE was mutated to LxGxK, did purpose, we used a well-characterized MCF-7cell not protect cells from cell death and resulted in an system in which exposure of cells to UV-C light results increase in cell death (compare column 3 with column in cell death by apoptosis (Pennaneach et al., 2001). As 2). Together these observations indicated that LxCxE shown in Figure 9a, overexpression of p202 in MCF-7 motif-dependent interaction of p202 to Rb (and other cells signiﬁcantly (40%) reduced apoptosis induced pocket proteins) increases cell survival in response to after exposure to UV-C light (compare column 2 with exposure of MCF-7cells to UV light.

Discussion

In murine fibroblast cell lines (AKR-2B and NIH 3T3) levels of p202 increased under reduced serum conditions and the addition of growth factors, such as PDGF, bFGF or TGF-b1, abrogated the increase in p202 levels (Geng et al., 2000). These observations raised the possibility that the signaling pathways, which are activated by cell growth factors receptors, negatively regulate expression of Ifi202-gene. Our observations reported here provide a molecular basis for decreases in p202 levels after the addition of cell growth factors to culture medium under reduced serum conditions. We suggest that phosphorylation of Rb by cyclin-dependent kinases after addition of cell growth factors inhibits the ability of Rb to cooperate with JunD to upregulate the expression of p202. Similarly, expression of SV40 large T antigen or adenovirus encoded E1A protein, which bind to Rb (Nevins, 1998), also impair the ability of Rb to upregulate p202 expression (see Figure 2). Con- versely, the signaling pathways that inhibit phosphorylation of Rb are predicted to upregulate p202 expression in a Rb-dependent manner. Consistent with this prediction, we note that IFN-induced levels of p202 were much higher in wild-type cells than Rb-null cells (Figure 1). Similarly, ectopic expression of p16INK4a,a positive regulator of Rb (Weinberg, 1995), strongly stimulated the activity of 202-luc-reporter in promoter– reporter assays (see Figure 5). Oncogenic transformation of NIH 3T3 cells by activated H-Ras (Q61L) results in upregulation of 202 Figure 9 p202 cooperates with Rb to increase cell survival. (a) mRNA and protein (Xin et al., 2003). Moreover, Cells of MCF-7human breast cancer cell line were transfected with increases in p202 levels in these cells correlate with pCMV, pCMV-202S or pCMV-202AS plasmid along with pCMV- increase in c-Jun/AP-1 DNA binding activity. Impor- b-gal plasmid (in 10 : 1 ratio). Transfected cells were exposed to UV tantly, c-Jun/AP-1 complex could bind to an AP-1 DNA light and cells were further incubated for 17–24 h. Cells were fixed 0 and stained for b-galactosidase activity. Blue cells were counted in binding sites (the 202AP-1CS1) present in the 5 - 4300 fields in each transfection. Bars indicate the standard regulatory region of the 202-gene in gel-shift mobility deviation. (b) MCF-7cells were transfected with pCMV, pCMV- assays and mutations in this site significantly decreased 202 or pCMV-202LxGxK plasmid along with pCMV-b-gal plasmid the activity of 202-luc-reporter in promoter–reporter (in 10 : 1 ratio). Transfected cells were exposed to UV light and cells assays (Xin et al., 2003). Our observations reported here were further incubated for 17–24 h. Cells were fixed and stained for b-galactosidase activity. Blue cells were counted in 4300 fields in indicate that Rb cooperates with JunD/AP-1 to activate each transfection. Bars indicate the standard deviation the transcription of Ifi202 gene. Together, these

Oncogene Rb upregulates p202 expression HXinet al 4782 observations support the idea that under reduced serum of p202 by Rb in MEFs and myoblasts contributes to conditions, when Rb is functional and levels of JunD are increased cell survival under stress conditions. In the relatively high (levels of c-Jun and JunB are relatively later case, it results in potentiation of differentiation of low), Rb recruits JunD/AP-1 to activate transcription of myoblasts to myotubes by p202 (Yee et al., 1998). Ifi202 gene through the AP-1 DNA binding site in the Others and we have previously expressed p202 in Ifi202-promoter. However, under conditions, such as in human cells to study p202-mediated cell growth effects transformed NIH 3T3 cells, when levels of JunD are low (Choubey and Kotzin, 2002). However, we want to and levels of c-Jun are much higher (Pfarr et al., 1994), point out that p202 does not have a direct structural c-Jun can activate transcription of 202 gene through the homologue in human cells. Moreover, human cells same AP-1 DNA binding site (independent of Rb). express structurally related proteins, such as IFI 16, Regulation of 202-promoter by both c-Jun and JunD is MNDA and AIM2, from the p200-protein family. It not new because c-Jun and JunD both can activate c- remains to be seen whether expression of mouse p202 in myb promoter via an AP-1-like element (Nicolaides human cells affects expression and/or functions of these et al., 1992). As ectopic expression of p202 inhibits c- proteins. Jun/AP-1-mediated transcription (Min et al., 1996), our Expression of JunD is known to increase cell survival observations also support the idea that c-Jun/AP-1 and by inhibiting p53-dependent apoptosis (Weitzman et al., p202 are part of a negative feedback loop, which may be 2000). As p202 inhibits p53-mediated transcription important for the regulation of cell growth. (Datta et al., 1996), upregulation of p202 by JunD (in Rb is also known to activate transcription of genes by cooperation with Rb; see Figure 7a) provides a novel stimulating the transcriptional activity of Sp1 (Udvadia mechanism by which increased levels of p202 inhibit et al., 1993). Interestingly, the 50-regulatory region of p53-induced apoptosis. Ifi202 gene contains a potential DNA-binding site for IFN-g treatment of carcinoma cell lines results in the transcription factor Sp1. Therefore, further work apoptosis (Berry et al., 1996). However, expression of will be needed to determine whether Rb could also Rb in these cell lines inhibits apoptosis induced by IFN- activate transcription of Ifi202 gene under certain g. Our observation that Rb upregulates p202 expression conditions. and p202 cooperates with Rb to increase cell survival The Rb tumor suppressor is best known for its role in also provides a molecular basis by which increased levels the regulation of cell cycle progression through its of p202 contribute to modulation of apoptosis by IFNs. binding with the transcription factor E2F (Nevins, 1998; The Rb protein is an essential mediator of IFN- Harbour and Dean, 2000). However, naturally occur- mediated inhibition of cell growth (Grander et al., 1997). ring Rb mutation (mutant R661W) has been identified As IFN-inducible proteins mediate the biological that abolishes E2F binding but not growth suppression activities of IFNs (Sen, 2000), our observation that (Kratzke et al., 1994; Otterson et al., 1997). Carriers of p202 cooperates with Rb to inhibit colony formation this mutation have only a 10% chance (100% chance in (Figure 8) also supports the idea that p202 mediates the Rb-null carriers) of developing Rb (DiCiommo et al., growth-inhibitory activities of IFNs, in part, through 2000), providing genetic evidence that Rb retains Rb. Similarly, p204, an another member of IFN- significant tumor suppression activities in the absence inducible p200-protein family, also inhibits cell growth of its E2F binding. As p202 binds to Rb and expression in Rb-dependent manner (Hertel et al., 2000). of p202 can inhibit E2F-mediated transcription (inde- Differential post-translational modification of Rb and pendent of Rb) (Choubey et al., 1996; Choubey and its degradation by a caspase in splenocytes derived from Gutterman, 1997), our observations described here C57BL/6 strain of mice, but not in splenocytes derived provide a novel mechanism by which wild-type Rb, by from DBA/2 strain of mice, is linked to increased upregulating p202 levels and cooperating with it (see sensitivity to radiation-induced apoptosis (Wallace et al., Figure 8), can inhibit E2F-mediated transcription 2001). As expression of p202 is not detectable in independent of E2F binding. The inhibition of E2F splenocytes derived from C57BL/6 strain of mice (p202 activity is known to result in inhibition of cell growth is expressed in splenocytes derived from DBA/2 strain of (Nevins, 1998). mice) (Choubey and Kotzin, 2002), our observations Loss of Rb gene results in embryonic lethality because reported here raise the possibility that the lack of p202 of excessive cell death by apoptosis (Weinberg, 1995) expression in splenocytes derived from C57BL/6 strain and MEFs derived from Rb-null embryo are susceptible of mice contributes to increased susceptibility to to apoptosis under reduced serum conditions (Almasan radiation-induced apoptosis. Further work will be et al., 1995). Additionally, Rb functions to inhibit needed to determine how mice strain-specific post- apoptosis during myocyte differentiation (Wang et al., translational modifications in Rb affect transcriptional 1997). Moreover, ectopic expression of Rb in a variety activation of Ifi202 gene in C57BL/6 strain of mice. of cell systems is known to provide resistance to In summary, our observations provide a molecular apoptosis induced by stress stimuli (Tan and Wang, basis by which expression of Ifi202 gene is regulated by 1998). Together, these observations support the idea Rb, JunD/AP-1 and their regulators under physiological that Rb supports cell survival under stress conditions. conditions. More importantly, our observations support Our observation that functional p202 cooperates with a model (see Figure 10) in which Rb and JunD, the two Rb to increase cell survival in MCF-7cells (see Figure 9) key regulators of cell growth arrest and cell survival provides a potential mechanisms by which upregulation (Weinberg, 1995; Karin et al., 1997), by upregulating

Oncogene Rb upregulates p202 expression HXinet al 4783 RB(L769D). Mutations in the 202AP-1 DNA binding sites (the 202AP-1CS1) were introduced using QuikChange site- directed mutagenesis kit from Stratagene (La Jolla, CA, USA) as suggested by supplier. The 202AP-1CS1 site (TGACTAA) was changed to AACCTAA. The pRL-TK reporter vector allowing expression of Renilla luciferase was purchased from Promega (Madison, WI, USA).

Colony-formation assays Colony-formation assays were done essentially as described previously (Xin et al., 2001). In brief, cells of Saos-2 cell line (B105 cells in 60-mm plate) were transfected with either empty vector (pCMV; 2 mg) or equal amounts of pCMV mRb or pCMV-202 plasmid, using calcium-phosphate transfection kit from Gibco-BRL. Following 24 h of transfections, cells were split into two 100-mm plates and selected in G418 for about 2 weeks. Colonies (0.5–1 mm in size) exhibiting G418-resistance were stained with crystal violet and counted as described previously (Xin et al., 2001).

Cell survival assays Figure 10 A hypothetical model indicating the regulation of p202 These assays were performed essentially as described before expression by the physiological regulators of Rb and JunD/AP-1 (Pennaneach et al., 2001). In brief, MCF-7cells were and cooperation of p202 with Rb to inhibit cell growth and increase cell survival transfected (using Lipofectamine-2000 reagent, Gibco-BRL) with plasmid allowing expression of 202 mRNA (pCMV-202), antisense to 202 mRNA (pCMV-202AS) or a mutant of 202 (pCDNA3.1-202LcGxK) along with a vector (pCDNA3-b-gal), p202 expression, inhibit cell growth and increase cell allowing b-galactosidase expression, in 10 : 1 ratio. DNA survival under certain stress conditions. damage was induced 24–36 h after transfection and cells were further incubated for 17–24 h in fresh medium. UV irradiation was performed with UV Stratalinker (STRATAGENE, Inc.). The cells were fixed with 0.1% glutaraldehyde and incubated Materials and methods overnight in the presence of X-gal (1 mg/ml) to detect b- galactosidase activity. The relative number of surviving cells Cell cultures after irradiation in the presence of the appropriate transfected NIH 3T3, Saos-2 and MCF-7cells were purchased from plasmid was determined by counting the number of blue cells American Type Culture Collection. Drs M Classon and Ed in 4300 nonoverlapping fields in unirradiated control and Harlow (Harvard Medical School, Boston, MA, USA) UV-irradiated cells for each transfection. generously provided MEF 3T3 cell lines derived from wild- þ / þ À/À type (Rb ) and Rb-null (Rb ) embryo. Drs J Weitzman Flow cytometric analysis and M Yaniv (Pasteur Institute, Paris) provided MEFs derived from wild-type (JunD þ / þ ) and JunD-null (JunDÀ/À) embryo. Flow cytometry was performed on single-cell suspensions on Cells were maintained at low density in Dulbecco’s modified adherent (after trypsin and EDTA treatment) as well as the Eagle’s medium (with high glucose) (DMEM) supplemented floating cells after pooling them. Briefly, for cell cycle analysis, with either 10% fetal bovine serum or 10% calf serum (for cells were stained with propidium iodide (50 mg/ml, Sigma) and MEF cell lines) and antibiotics in an incubator with 5% CO2. subjected to flow cytometry using Coulter Epics XL-MCL flow If so indicated, cells were treated with either recombinant IFN cytometer as described previously (D’Souza et al., 2001). (1000 U/ml) for the indicated time. For serum starvation, cells were washed with PBS and incubated with medium supplemented with indicated concentration of serum for 36 h. Immunoblotting Cells were collected from plates in PBS and resuspended in a modified RIPA lysis buffer (50 mm Tris-HCl, pH 8.0; 250 mm Plasmids NaCl, 1% NP-40, 0.5% sodium deoxycholate, 0.1% SDS) The 202-luc-reporter plasmid, containing the 50-flanking supplemented with protease inhibitors (leupeptin, 50 mg/ml; sequence (0.8 kb) from the Ifi202 gene has been described pepstatin A, 50 mg/ml; PMSF, 1 mm), and incubated at 41C for previously (D’Souza et al., 2001). Dr Paul Hamel (University 30 min. The cell lysates were sonicated briefly before centrifu- of Toronto, Canada) generously provided the plasmid encod- gation at 14 000 r.p.m. in a microfuge for 10 min. The ing a mutant form of Rb (Dp34), defective in phosphorylation supernatants were collected and equal amounts of proteins by cyclin-dependent kinase; see Hamel et al., 1992). Dr Sunita were processed for immunoblotting as described previously Agarwal (National Institute of Diabetes and Digestive and (Choubey and Lengyel, 1993). The p202 polyclonal antiserum Kidney Diseases, Bethesda, MD, USA) generously provided has been described previously (Choubey and Lengyel, 1993). mouse JunD cDNA in plasmid pcDNA3.1. Dr Jean Y Wang Antibodies to JunD, cyclin D1, E2F1 and actin were (University of California, San Diego, La Jolla, CA, USA) purchased from Santa Cruz Biotech. Inc. (Santa Cruz, CA, generously provided Rb pocket mutants Rb(N757F) and USA).

Oncogene Rb upregulates p202 expression HXinet al 4784 Northern blotting Luciferase assays Total cytoplasmic RNA was isolated from cells and subjected Reporter assays were performed as described previously to Northern blotting followed by hybridization with a HincII (D’Souza et al., 2001). In brief, subconfluent cells in six-well fragment of the 202 cDNA (Choubey et al., 1989). Hybridiza- plates were transfected with the reporter plasmids 202-luc tions were performed as described previously (D’Souza et al., (5 mg) and pRL-TK (0.5 mg) using the calcium phosphate 2001). transfection system (Gibco-BRL Life Tech., Rockville, MD, USA), as suggested by the supplier. Cells were harvested 42– Reverse transcriptase–polymerase chain reaction (RT–PCR) 48 h after transfections and the firefly luciferase and Renilla luciferase activities were determined as described previously. In To study the expression of the 202 gene, total RNA was reporter assays, shown in Figures 4c, 7a and b, cells were isolated from cells and subjected to the 202a-specific RT-PCR incubated in the culture medium supplemented with 0.5% 0 using the Ifi/202-gene specific forward (5 -GGT CA TCTACC serum for 24 h before harvesting for luciferase activity. The 0 0 AACTCA GAAT-3 ) and backward primer (5 -TCTAG- luciferase activity in control vector transfected cells is shown 0 GATGCCACTGCTGTTG-3 ) and Superscript One-Step as 1. RT–PCR System from Gibco-BRL (Grand Island, NY, USA). As a control, to determine the amounts of input RNA in the above RT–PCR reactions, b-actin-specific primers Acknowledgements from Gibco-BRL (Grand Island, NY, USA) were used for We thank Drs Qin, Geng, and D’Souza for their help during RT–PCR. The 202-specific (358-bp) and b-actin specific (353- the initial experiments. We also thank Dr Peter Lengyel for bp) RT–PCR products were analysed by agarose gel electro- discussions and suggestions. This work was supported, in part, phoresis. by a grant from the National Institutes of Health (CA69031).

References

Almasan A, Yin Y, Kelly RE, Lee EY, Bradley A, Li W, D’Souza S, Xin H, Walter S and Choubey D. (2001). J. Biol. Bertino JR and Wahl GM. (1995). Proc. Natl. Acad. Sci. Chem., 276, 298–305. USA, 92, 5436–5440. Geng Y, D’Souza S, Xin H, Walter S and Choubey D. (2000). Berry DE, Lu Y, Schmidt B, Fallen PG, O’Connell C, Cell Growth Diff., 11, 475–483. Hu SX, Xu HJ and Blanck G. (1996). Oncogene, 12, 1809– Grander D, Sangfelt O and Erickson S. (1997). Eur. J. 1819. Haematol., 59, 129–135. Castellazzi M, Spyrou G, La Vista N, Dangy JP, Piu F, Yaniv Gribaudo G, Toniato E, Engel DA and Lengyel P. (1987). J. M and Brun G. (1991). Proc. Natl. Acad. Sci. USA, 88, Biol. Chem., 262, 11878–11883. 8890–8894. Gutterman JU and Choubey D. (1999). Cell Growth Diff., 10, Charles A, Tang X, Crouch E, Brody JS and Xiao ZX. (2001). 93–100. J. Cell. Biochem., 83, 414–425. Hamel PA, Gill RM, Phillips RA and Gallie BL. (1992). Chen PL, Riley DJ, Chen Y and Lee WH. (1996a). Genes Dev., Oncogene, 7, 693–701. 10, 2794–2804. Haput Y, Rowan S and Oren M. (1995). Oncogene, 10, 1563– Chen PL, Riley DJ, Chen KS and Lee WH. (1996b). Proc. 1571. Natl. Acad. Sci. USA, 93, 465–469. Harbour JW and Dean DC. (2000). Genes Dev., 14, 2393–2409. Chen TT and Wang JY. (2000). Mol. Cell. Biol., 20, Hertel L, Rolle S, De Andrea M, Azzimonti B, Osello R, 5571–5580. Gribaudo G, Gariglio M and Landolfo S. (2000). Oncogene, Choubey D. (2000). J. Biol. Regul. Homeost. Agents, 14, 19, 3598–3608. 187–192. Huang HJ, Yee JK, Shew JY, Chen PL, Bookstein R, Choubey D and Gutterman JU. (1997). Oncogene, 15, Friedmann T, Lee EY and Lee WH. (1988). Science, 242, 291–302. 1563–1566. Choubey D and Kotzin BL. (2002). Front Biosc., 7, e252–e262. Johnstone RW and Trapani JA. (1999). Mol. Cell. Biol., 19, Choubey D and Lengyel P. (1993). J. Interferon Res., 13, 5833–5838. 43–52. Karin M. (1995). J. Biol. Chem., 271, 16483–16486. Choubey D and Lengyel P. (1995). J. Biol. Chem., 270, 6134– Karin M, Liu ZG and Zandi E. (1997). Curr. Opin. Cell Biol., 6140. 9, 240–246. Choubey D, Li SJ, Datta B, Gutterman JU and Lengyel P. Kim SJ, Lee HD, Robbins PD, Busam K, Sporn MB and (1996). EMBO J., 15, 5668–5678. Roberts AB. (1991). Proc. Natl. Acad. Sci. USA, 88, 3052– Choubey D, Snoddy J, Chaturvedi V, Toniato E, Opdenakker 3056. G, Thakur A, Samantha H, Engel D and Lengyel P. (1989). Kratzke RA, Otterson GA, Hogg A, Coxon AB, Geradts J, J. Biol. Chem., 264, 17182–17189. Cowell JK and Kaye FJ. (1994). Oncogene, 9, 1321–1326. Datta B, Li B, Choubey D, Nallur G and Lengyel P. (1996). J. Lembo D, Angeretti A, Benefazio S, Hertel L, Gariglio M, Biol. Chem., 271, 27544–27557. Novelli F and Landolfo S. (1995). J. Biol. Regul. Homeost. Datta B, Min W, Burma S and Lengyel P. (1998). Mol. Cell. Agents, 9, 42–46. Biol., 18, 1074–1083. Li H and Wicks WD. (2001). Arch. Biochem. Biophys., 394, Decary S, Decesse JT, Ogryzko V, Reed JC, Naguibneva I, 1–12. Harel-Bellan A and Cremisi CE. (2002). Mol. Cell. Biol., 22, Lipinski MM and Jacks T. (1999). Oncogene, 18, 7873–7882. 7877–7888. Min W, Ghosh S and Lengyel P. (1996). Mol. Cell. Biol., 16, Decesse JT, Medjkane S, Datto MB and Cremisi CE. (2001). 359–368. Oncogene, 20, 962–971. Nead MA, Baglia LA, Antinore MJ, Ludlow JW and DiCiommo D, Gallie BL and Bremner R. (2000). Cancer Biol., McCance DJ. (1998). EMBO J., 17, 2342–2352. 10, 255–269. Nevins JR. (1998). Cell Growth Differ., 9, 585–593.

Oncogene Rb upregulates p202 expression HXinet al 4785 Nicolaides NC, Correa I, Casadevall C, Travali S, Soprano KJ Udvadia AJ, Rogers KT, Higgins PD, Murata Y, Martin KH, and Calabretta B. (1992). J. Biol. Chem., 267, 19665–19672. Humphrey PA and Horowitz JM. (1993). Proc. Natl. Acad. Nishitani J, Nishinaka T, Cheng CH, Rong W, Yokoyama Sci. USA, 90, 3265–3269. KK and Chiu R. (1999). J. Biol. Chem., 274, 5454–5461. Wallace M, Coates PJ, Wright EG and Ball KL. (2001). Otterson GA, Chen WD, Coxon AB, Khleif SN and Kaye FJ. Oncogene, 20, 3597–3608. (1997). Proc. Natl. Acad. Sci. USA, 94, 12036–12040. Wang JY. (1997). Curr. Opin. Genet. Dev., 7, 39–45. Pennaneach V, Salles-Passador I, Munshi A, Brickner H, Wang J, Guo K, Wills KN and Walsh K. (1997). Cancer Res., Regazzoni K, Dick F, Dyson N, Chen TT, Wang JY, 57, 351–354. Fotedar R and Fotedar A. (2001). Mol. Cell, 7, 715–727. Weinberg RA. (1995). Cell, 81, 323–330. Pfarr CM, Mechta F, Spyrou G, Lallemand D, Carillo S and Weitzman JB, Fiette L, Matsuo K and Yaniv M. (2000). Mol. Yaniv M. (1994). Cell, 76, 747–760. Cell, 6, 1109–1119. Rustgi AK, Dyson N and Bernards R. (1991). Nature, 352, Xin H, D’Souza S, Lei F, Lengyel P and Choubey D. (2001). 541–544. Oncogene, 20, 6828–6839. Samanta H, Engel DA, Chao HM, Thakur A, Garcia-Blanco Xin H, Geng Y, Pramanik R and Choubey D. (2003). J. Cell. MA and Lengyel P. (1986). J. Biol. Chem., 261, 11849–11858. Biochem., 88, 191–204. Sen GC. (2000). Semin. Cancer Biol., 10, 93–101. Yee AS, Shin HH and Tevosian SG. (1998). Front. Biosci., 3, Tan X and Wang JY. (1998). Trends Cell Biol., 8, D532–D547. 116–120. Zheng L and Lee WH. (2001). Exp. Cell Res., 264, 2–18.

Oncogene