BTG2TIS21/PC3 Induces Neuronal Differentiation and Prevents

BTG2TIS21/PC3 induces neuronal diﬀerentiation and prevents apoptosis of terminally diﬀerentiated PC12 cells

Fatiha el-Ghissassi1, Sandrine Valsesia-Wittmann1, Nicole Falette1, Cyril Duriez1, Paul D Walden2,3 and Alain Puisieux*,1,4

1De´partement d’Oncologie Fondamentale et Appliqueé, INSERM Unite´ 453. Centre León Be´rard, 28 rue Lae¨nnec, 69008, Lyon, France; 2Department of Biochemistry, NYU School of Medicine, 550 First Avenue, New York, NY 10016, USA; 3Department of Urology, NYU School of Medicine, 550 First Avenue, New York, NY 10016, USA; 4De´partement de Geńe´tique Molećulaire et Biochimie Clinique, Faculte´ de Pharmacie, 69008, Lyon, France

The p53-transcriptional target, BTG2TIS21/PC3, was enhanced in response to genotoxic stress through a p53- previously identified as an antiproliferative gene. However, dependent mechanism (Rouault et al., 1996; Cortes et al., the precise biological functions of the protein product 2000). BTG2TIS21/PC3 is also transiently expressed in remain to be elucidated. BTG2TIS21/PC3 expression is response to signals such as epidermal growth factor induced in vivo during neurogenesis, and the gene is (EGF), tumor promoter agent (TPA), or the addition of transiently expressed in vitro in rat pheochromocytoma serum to starved cells (Bradbury et al., 1991; Fletcher et PC12 cells after induction of neuronal differentiation by al., 1991). However, several in vivo and in vitro addition of nerve growth factor (NGF). These observations observations suggest that it may play a role in neuronal suggest that BTG2TIS21/PC3 is functionally significant differentiation. In vivo, BTG2TIS21/PC3 is considered as a during the neuronal differentiation process. To test this marker of neuronal birth (Iacopetti et al., 1994). Indeed, hypothesis, a vector that expressed BTG2TIS21/PC3 under the BTG2TIS21/PC3 gene is specifically expressed in the control of an inducible promoter was introduced into neuroepithelial cells that will generate postmitotic PC12 cells. Growth arrest and differentiation in response neurons (Iacopetti et al., 1999). Its overexpression affects to NGF were greatly enhanced by BTG2TIS21/PC3 over- the pattern of cell division of cortical precursors in rats expression. Furthermore, an antisense oligonucleotide (Malatesta et al., 2000). In vitro, BTG2TIS21/PC3 is complementary to BTG2TIS21/PC3 mRNA, which was able activated by NGF at the onset of neuronal differentiation to inhibit endogenous BTG2TIS21/PC3 expression, triggered in PC12, a cell line derived from a murine adrenal medulla programmed cell death in differentiated PC12 cells. These tumor (Bradbury et al., 1991). In the PC12nn25 mutant observations confirm that BTG2TIS21/PC3 expression cell line, characterized by a defect in NGF receptor- promotes neuronal differentiation and that it is required mediated endocytosis, BTG2TIS21/PC3 response to NGF is for survival of terminally differentiated cells. completely abolished and cells remain undifferentiated Oncogene (2002) 21, 6772 – 6778. doi:10.1038/sj.onc. (Altin et al., 1991). Interestingly, the BTG2TIS21/PC3 1205888 protein physically interacts with and activates the PRMT1 protein-arginine-N-methyltransferase (Lin et Keywords: BTG2TIS21/PC3; PC12 cells; cell differentia- al., 1996) whose function is required for PC12 differentiation; apoptosis tion by NGF (Cimato et al., 1997). To directly investigate the role of BTG2TIS21/PC3 in neuronal differentiation, rat pheochromocytoma PC12 cells were transfected with an Introduction inducible BTG2TIS21/PC3 expression vector. We show that BTG2TIS21/PC3 overexpression enhances NGF-induced The human BTG2TIS21/PC3 gene is a member of a newly neuronal differentiation of PC12 cells. Inhibition of identified family of six independent antiproliferative BTG2TIS21/PC3 mRNA with specific antisense oligonu- genes, namely, BTG1, ANA/BTG3, PC3B, TOB, TOB2 cleotide promotes programmed cell death in NGF (Rouault et al., 1992; Matsuda et al., 1996; Guehenneux et differentiated cells and in cells overexpressing al., 1997; Yoshida et al., 1998; Ikematsu et al., 1999; BTG2TIS21/PC3. We therefore conclude that the over- Buanne et al., 2000; Tirone, 2001). Although the expression of BTG2TIS21/PC3 is required for the survival of biological functions of the BTG2TIS21/PC3 protein remain differentiated PC12. to be elucidated, several studies have attempted to identify the different cues capable of inducing its expression. We Results previously reported that BTG2TIS21/PC3 expression is Generating and characterizing stable PC12 lines containing an inducible BTG2TIS21/PC3 construct *Correspondence: A Puisieux, Centre Le´ on Be´ rard, 28 rue Lae¨ nnec, 69008, Lyon, France; E-mail: [email protected] Clonal PC12 cell lines were generated with TIS21/PC3 Received 29 April 2002; revised 4 July 2002; accepted 18 July 2002 BTG2 cDNA under the control of the Lac BTG2TIS21/PC3 triggers differentiation of PC12 cells F el-Ghissassi et al 6773 repressor protein. Expression was induced by adding addition, then lasted 6 days after IPTG was removed iso propyl-b-D-thiogalactoside (IPTG), a non toxic from the medium (results not shown). lactose analog, to the culture medium. Using the Lac switch system, it was possible to generate stable cell Growth of PC12 cells containing an inducible lines in which BTG2TIS21/PC3 expression was induced BTG2TIS21/PC3 construct following the addition of IPTG to the cell culture media. The 1100 bp fragment which was used The proliferative status of PC12-18 and PC12-27 cells contained the entire coding region but lacked the cultured in the absence and in the presence of IPTG 3’untranslated region, making it possible to discrimi- was investigated. Similar data were obtained for both nate it from the endogenous messenger. Forty-three clones. As shown on Figure 2c, IPTG-induced different clonal lines were screened. In all of them, BTG2TIS21/PC3 expression caused a significant decrease as expected, endogenous BTG2TIS21/PC3 mRNA in cell growth whereas neither cell viability nor cell was almost undetectable. Ten showed inducible morphogenesis were significantly affected (data not BTG2TIS21/PC3 expression after IPTG addition. Three sublines were selected for further analysis (Figure 1): PC12718 and PC12-27 exhibiting high inducible BTG2TIS21/PC3 expression and as a control PC12-2 clone, which displayed no expression of BTG2TIS21/PC3, even after IPTG addition. As shown on Figure 2a, the expression of exogenous BTG2TIS21/PC3 mRNA was increased in PC12-18 and PC12-27 cells following exposure to IPTG for 12 h. BTG2TIS21/PC3 protein levels also increased after IPTG addition, with a kinetic profile similar to that of exogenous BTG2TIS21/PC3 mRNA (Figure 2b). Maximum expression levels were reached within 3 days of IPTG exposure. In PC12 sublines, the ectopic expression persisted up to 8 days following IPTG

Figure 2 Characterization of inducible BTG2TIS21/PC3 expression Figure 1 Characterization of stable PC12 sublines containing in- and effect on cell growth. (a) Northern blot analysis of IPTG-inducible BTG2TIS21/PC3 expression. Northern blot analysis of ecto- duced BTG2TIS21/PC3 mRNA expression in PC12 cells (clone pic BTG2TIS21/PC3 mRNA in PC12 cells (clone PC12-18 and clone PC12-27) stably transfected with an inducible BTG2TIS21/PC3 ex- PC12-27) stably transfected with an inducible BTG2TIS21/PC3 expression vector. Ten mg of total cell RNA were analysed after pression vector. Ten mg of total RNA were analysed after 12-h in- 0.5, 1, 2 and 3 days incubation with 3 mM IPTG. (b) Western blot TIS21/PC3 cubation with 3 mM IPTG. Clone PC12-2 is a negative control analysis of ectopic BTG2 protein expression in PC12 cells clone exhibiting no ectopic BTG2TIS21/PC3 expression even after (clone PC12-27) under the same conditions as (a). One hundred IPTG addition. The band at approximately 1.1 kb represents mg of total cell proteins were separated by SDS – PAGE and transfected BTG2TIS21/PC3 (T-BTG2TIS21/PC3), while the band at probed with BTG2TIS21/PC3 antibody. (c) Proliferation of cells fol- approximately 2.5 kb represents endogenous BTG2TIS21/PC3 lowing BTG2TIS21/PC3 induction was studied in PC12-27 cell line, TIS21/PC3 TIS21/PC3 mRNA (E-BTG2 ). Transfected BTG2 lacks ap- either untreated (green box), or following exposure to 3 mM proximately 1400 base pairs of the 3’untranslated region con- IPTG (pink box). Quantiﬁcation was performed by counting the tained in endogenous BTG2TIS21/PC3. 18S ribosomal RNA is cell number at a given time. Data are expressed as means (+stan- shown to demonstrate equal loading of total RNA dard deviation)

Oncogene BTG2TIS21/PC3 triggers differentiation of PC12 cells F el-Ghissassi et al 6774 shown). Cell cycle progression was then analysed using Discussion flow cytometry to measure changes in the DNA content. BTG2TIS21/PC3 overexpression in PC12-27 cells To understand the different aspects of neuronal caused partial growth inhibition within 1 day of IPTG differentiation, the PC12 rat pheochromocytoma cell exposure, with a slight decrease in the number of cells line has frequently been used as an in vitro model. in S phase (8 versus 13% in control cells) and a slight Undifferentiated PC12 cells proliferate in the presence increase in the number of cells in G1 (63 and 58% in of serum; however, the addition of NGF results in a induced and control cells, respectively). sympathetic-like neuronal differentiation, including complete mitotic arrest (Rudkin et al., 1989; van Grunsven et al., 1996). The present study demonstrates Effects of BTG2TIS21/PC3 expression on PC12 that the forced expression of the antiproliferative differentiation BTG2TIS21/PC3 gene enhances NGF-induced neuronal It is known that BTG2TIS21/PC3 is expressed in PC12 differentiation and favors the survival of differentiated cells after induction of neuronal differentiation PC12 cells. Cell-cycle arrest is known to be an essential by NGF. In order to demonstrate the role of preliminary requirement for terminal differentiation. BTG2TIS21/PC3 in NGF-induced differentiation, PC12- Other studies have shown that proliferation inhibitors 18 and PC12-27 lines were treated, either with NGF combined with NGF treatment, enhance the differ- alone or with a combination of IPTG and NGF entiation rate of PC12 cells (Gupta et al., 1987). (100 ng/ml). Neuronal differentiation was estimated by Similar findings resulted from culturing PC12 cells in neurite outgrowth which was correlated with enhanced serum-free medium prior to NGF addition (Rudkin et levels of the neuron-specific marker of differentiation a- al., 1989). However, the in vivo mechanisms governing enolase (data not shown) (Vinores et al., 1981; Marz et this process are not completely understood. Much al., 1997). After NGF treatment 20% of PC12-18 and attention has focused on the regulation of components PC12-27 lines developed neurites, whereas after known to control progression through cell cycle, IPTG/NGF treatment, up to 80% of cells differen- including cyclins, cyclin-dependent kinases and cyclin- tiated (Figure 3a,b). These data suggested that dependent kinase inhibitors (Erhardt and Pittman, induction of BTG2TIS21/PC3 expression, substantially 1998). In addition to previous reports, our data show enhanced responsiveness to NGF-induced neurite that BTG2TIS21/PC3 expression may play a significant outgrowth suggesting that these cells differentiated role on the control of cell growth. In vitro, more rapidly in the presence of BTG2TIS21/PC3. BTG2TIS21/PC3 was known to trigger a G1 arrest in pheochromocytoma PC12 cells and in non-neuronal cells (Montagnoli et al., 1996; Guardavaccaro et al., BTG2TIS21/PC3 is involved in the survival of NGF treated 2000). However, the inducible overexpression of PC12 BTG2TIS21/PC3 led only to a partial G1-arrest in the To confirm the role of BTG2TIS21/PC3 in the differ- absence of NGF (Figure 2), suggesting that its entiation process, differentiation of NGF-treated PC12 induction was not sufficient to trigger the G1 block cells was triggered in the presence of an antisense that characterizes neuronal differentiation by NGF. In phosphorothioate oligodeoxynucleotide designed to vivo, BTG2TIS21/PC3 expression was observed during interfere with BTG2TIS21/PC3 expression. After treat- neurogenesis, in the subset of neuroblasts differentiat- ment with antisense oligonucleotide AS-BTG2, the ing into post-mitotic neurons (Iacopetti et al., 1999). expression of the BTG2TIS21/PC3 protein was reduced to The expression of the protein was sustained during the levels lower than the basal level seen in untreated cells, mitosis of neuron-generating neuroepithelial cells, and as judged by Western blot analysis (Figure 4a). during a few days in post-mitotic neuronal daughter Conversely, BTG2TIS21/PC3 protein levels were unaf- cells. This observation is consistent with the hypothesis fected by treatment with a non specific oligonucleotide that BTG2TIS21/PC3 induction does not trigger a (Figure 4a). In the absence of NGF treatment, AS- complete cell-cycle arrest, but more probably reduces BTG2 had no effect on PC12 cells (Figure 4c). the cycling rate of neuroepithelial cells to below a However, the addition of NGF to PC12 cells treated certain threshold, thereby forcing the cells to switch by AS-BTG2 triggered a massive cell death, as from a proliferative symmetric mode (that generates evidenced by an increased number of floating cells. two proliferating daughter neuroepithelial cells) to The percentage of apoptotic cells was then evaluated asymmetric division (that gives one post-mitotic neuron by the TUNEL method: 20% of apoptotic cells were and one neuroepithelial cell) (Tirone, 2001; Malatesta observed 6 days after treatment with AS-BTG2, et al., 2000). This function may be common to compared to 4% after treatment with a control non other members of the BTG/TOB family. Like specific oligonucleotide (Figure 4b,c). This increase in BTG2TIS21/PC3, BTG3/ANA is expressed in the neuroe- apoptotic cell death was confirmed by cell-cycle pithelial cells of the ventricular zone of the developing analysis evidencing the emergence of a significant central nervous system (Yoshida et al., 1998). sub-G1 cell population (data not shown). Taken BTG2TIS21/PC3 and BTG3/ANA may thus have over- together these data strongly suggest that lapping roles in the growth arrest of neural precursor BTG2TIS21/PC3 is required for the survival of NGF cells at the time when the commitment of the precursor treated PC12 cells. cells to the neural or glial lineage occurs.

Oncogene BTG2TIS21/PC3 triggers differentiation of PC12 cells F el-Ghissassi et al 6775

Figure 3 Enhanced differentiation of PC12 cells after induction of BTG2TIS21/PC3 expression and NGF treatment. (a) Photomicro- graphs of undifferentiated and NGF-differentiated cells (clone 12-27) following 12 h pre-treatment with 3 mM IPTG (+ IPTG) or no treatment (control). (b) Measurement of neurite development of differentiating cells (clone PC12-27 + NGF) at 2, 3, 6 and 8 days, as well as in control and IPTG-treated cultures. The IPTG group was initially incubated with IPTG for 12 h prior to NGF treatment. In order to quantify differentiation discrepancies observed, cells with one or more neurite extensions of at least twice the cell body length were regarded as neurite positive. The number of cells with such outgrowth was counted from ﬁve randomly selected ﬁelds. Results are expressed as a percentage of the total number of cells. All cells on any individual dish were counted up to a total of 250 cells. Data represent means (+standard deviation) of three individual experiments. Representative photographs of cells were taken using an Axiophot2 camera attachment

Apart from the role of BTG2TIS21/PC3 as a signal factor in NGF treated cells, our team previously transducer at the onset of neurogenesis, data of showed that the differentiated embryonic stem cells experiments using antisense phosphorothioate oligo- from which BTG2TIS21/PC3 had been ablated underwent deoxynucleotide suggest that BTG2TIS21/PC3 expression apoptosis following DNA damage because of a failure is required for the survival of post-mitotic neuronal in growth arrest (Rouault et al., 1996). daughter cells. This observation is highly reminiscent of Another key event in the growth arrest associated the apoptotic cell death of differentiated PC12 cells with the differentiation process is the induction of the after NGF withdrawal (Greene, 1978). Similarly, the p21Waf1 cyclin-dependent kinase inhibitor. p21Waf1 is inhibition of BTG2TIS21/PC3 expression may trigger upregulated during the NGF-induced cell cycle block programmed cell death as a consequence of terminally in PC12 cells. The experimental overexpression of differentiated cells attempting to re-enter the cycle. p21Waf1 is sufficient to induce differentiation-specific Consistent with this role of BTG2TIS21/PC3 as a survival cell cycle events (Erhardt and Pittman, 1998). Interest-

Oncogene BTG2TIS21/PC3 triggers differentiation of PC12 cells F el-Ghissassi et al 6776 2000). p21Waf1 and BTG2TIS21/PC3 are transcriptionally induced by p73 (Zhu et al., 1998), their expression increases during the differentiation of neuroblastoma cells (unpublished data). Studies are ongoing to assess whether p21Waf1 and BTG2TIS21/PC3 act as p73 down- stream targets in this process. While this manuscript was under review, we learned that another, independent study reported that transient transfection of BTG2TIS21/PC3 was able to potentiate the effect of NGF on differentiation (Corrente et al., 2002). Combined data clearly demonstrate the role of BTG2TIS21/PC3 in neuronal differentiation.

Materials and methods

Cell culture

PC12 cells were grown at 378C (5% CO2) in DMEM (4.5 g/l glucose) with 6% donor heat-inactivated horse serum, 6% heat-inactivated fetal bovine serum, 10 mM HEPES, 10 mM glutamine, 100 units/ml penicillin and 100 mg/ml streptomycin. Cells were cultured in plastic tissue culture dishes pre-coated with rat-tail collagen (0.25 mg/ml) (Sigma) and poly-L-lysine (10 mg/ml) (Sigma). Growth medium was supplemented with 200 mg/ml hygromycin B (Gibco-BRL) and 300 mg/ml G418 (Gibco-BRL) in cell lines transfected with Lac repressor and BTG2TIS21/PC3 expression plasmids. For induction of TIS21/PC3 BTG2 in transfected lines, 3 mM IPTG (Boehringer Mannheim Biochemicals) were added to the medium. To induce diﬀerentiation, 100 ng/ml of nerve growth factor (NGF) (Sigma) were added to the same medium as above.

Plasmids For induction of BTG2TIS21/PC3 expression, plasmids from the Lac expression kit (Stratagene) were employed. Briefly, this system involves the constitutive expression of the Lac repressor protein and the introduction of a gene of interest (BTG2TIS21/PC3) under the control of a RSV promoter containing Lac repressor protein binding sites. Under normal Figure 4 Treatment with antisense oligonucleotide AS-BTG2 in- conditions, the Lac repressor protein binds to Lac operator hibits BTG2TIS21/PC3 protein expression and enhances cell death sites in the RSV promoter, thus inhibiting transcription. For TIS21/PC3 in NGF treated PC12 cells. PC12 cells were treated during 6 days induction of BTG2 , isopropyl-b-D-thiogalactoside with NGF in the absence of antisense oligonucleotide (control), in (IPTG) was added to the cells. IPTG binds to the Lac repressor the presence of control non specific oligonucleotide (NSO) or with protein, thus inhibiting interactions with DNA. The RSV the anti-BTG2TIS21/PC3 antisense oligonucleotide (AS-BTG2). promoter then initiates transcription, leading to ectopic Then, proteins were extracted and samples were analysed by Wes- TIS21/PC3 tern blotting (a). Detection and quantification of apoptotic cell BTG2 expression. The first plasmid, pCMV, contains were performed using the TUNEL staining method (b). Examples the Lac repressor protein coding sequence and the hygromicin of apoptotic bodies (arrows) in cells treated with BTG2TIS21/PC3 B resistance gene. The second plasmid, pOPRSV1, is an antisense oligonucleotides. (c) Quantification of apoptotic bodies inducible expression plasmid bearing geneticin G418 resistance, (values are means with standard deviation) in which a 1.1 kilobase human BTG2TIS21/PC3 cDNA fragment has been subcloned into the KpnI site. Both plasmids were cotransfected with lipofectin (Gibco-BRL) into PC12 cells ingly, both BTG2TIS21/PC3 and p21Waf1 are known to be according to the manufacturer’s instructions. The selection for transcriptional targets of the tumor suppressor p53, in the clonal expression of both the Lac repressor protein and response to DNA damage (Rouault et al., 1996; el- the BTG2TIS21/PC3 construct was obtained by addition of Deiry et al., 1993). In contrast to p53, the p53-homolog hygromicin (200 mg/ml) (Gibco-BRL) and geneticin G418 p73 plays a critical role in neuronal differentiation. (300 mg/ml) (Gibco-BRL). Lac repressor protein expression P73-deficient mice show major neurological defects was tested by Western blot analysis and the expression inducible in clones was confirmed with Northern blot and (Yang et al., 2000). In vitro, endogenous p73 levels Western blot analyses. increase in neuroblastoma cells induced to differentiate by retinoic acid. The experimental overexpression of p73 (but not p53) is sufficient to induce both Cell proliferation assay morphological (neurite outgrowth) and biochemical Cell growth was determined by the measurement of cell markers of neuronal differentiation (de Laurenzi et al., numbers after incubation with Trypan blue (0.06% in PBS).

Oncogene BTG2TIS21/PC3 triggers differentiation of PC12 cells F el-Ghissassi et al 6777 Unstained cells (living cells) were counted in three separate After centrifugation, the cell pellets were washed twice with wells per experiment; five random fields were counted in each PBS, then treated with 100 mg/ml DNase-free RNase A in well. The relative cell number is the ratio of the cell count at PBS for 30 min. The nuclei were then stained with propidium a given time to the number of cells at time 0. iodide used at the concentration of 50 mg/ml. The DNA content of the stained nuclei was measured on a FACScalibur flow cytometer. The Cell Quest software (Becton Dickinson, RNA isolation and Northern blot analysis USA) was used to determine cell cycle phase distribution. Total RNA from PC12 cells was isolated using the tri-reagent method (Sigma). RNA samples (10 mg) were electrophoresed Antisense oligonucleotides on denaturating formaldehyde agarose gel, then transferred to nylon membranes (Hybond-N+; Amersham). Membranes Phosphorothioate oligonucleotides (100 mM) and lipofectin were hybridized with a a-32P-dCTP-labeled BTG2TIS21/PC3 (1 mg/ml) (Gibco) were incubated for 15 min at 378C. The cDNA probe. Bands were visualized by autoradiography. oligonucleotide-lipofectin mixture was diluted with serum- containing medium, then added to the cells. In most cases, the dilution was 1 : 100, yielding a final oligonucleotide- Protein isolation and Western blot analysis concentration of 1 mM. Fresh oligonucleotide-containing Cells were lysed on ice in RIPA buffer (50 mM Tris-HCl, medium was added to the cells each day. We used 22-mer 0.1% SDS, 2 mM DTT, 0.5% NP40) containing 1 mM antisense oligonucleotides complementary to the region of the PMSF, 40 mg/ml leupeptin, 40 mg/ml aprotinin, 20 mg/ml BTG2TIS21/PC3 mRNA that encompasses the initiation codon. pepstatin A. Total protein concentrations were determined The oligonucleotides were purchased from GENSET by the Bio-Rad protein assay. Aliquots of each sample (France). A non specific phosphorothioate 22-mer oligonu- containing equal amounts of proteins were subjected to cleotide (NSO) was used as control. Western immunoblot analysis. Briefly, proteins were separated by electrophoresis on a 10% polyacrylamide gel, then Detection of apoptotic cell death transferred to Immobilon polyvinylidene difluoride transfer membrane (Millipore). The detection of apoptotic cell death was performed using the fluorescein-based in situ cell detection Kit (Enzo Roche). The assay was based upon the detection of DNA degradation Immunological studies using TUNEL (terminal dUTP-fluorescein nick end labeling) For detection by Western blot, equal amounts of proteins staining. The TUNEL assay was conducted on cytospin were mixed with Laemmli sample buffer, resolved on 12% preparations of cells removed from dishes by trituration in SDS-polyacrylamide gel, then transferred onto PVDF the culture medium (to avoid loss of detached cells) as per membranes (Millipore). Western blot analysis was performed manufacturer’s instructions. Apoptotic cells were visualized using enhanced chemiluminescence in accordance with the under a Zeiss axioplot 2 fluorescence microscope. The extent manufacturer’s instructions. Specific antibodies to the of apoptosis was expressed as the ratio of cells that showed BTG2TIS21/PC3 protein were generated by immunizing rabbits positive TUNEL staining on the total number of cells (the with six times His-tagged BTG2TIS21/PC3 protein, followed by total number of cells was calculated from cells stained with affinity purification involving adsorption to, and elution from Hoechst 33342 (10 mg/ml)). Typically, at least 300 cells per a column of BTG2 protein and by negative adsorption to a treatment were counted in random fields. Three independent column of BTG1 protein (Walden et al., 1998). For the experiments were performed in each study. detection of Neuron Specific aEnolase, the primary antibody was Enolase C-19 (Santa Cruz). Peroxidase-conjugated anti- rabbit or anti-goat IgG (Dako) were used as secondary antibodies. Acknowledgements We thank MD Reynaud and N Borel for their help in Cell cycle analysis using flow cytometry (FACS) preparing the manuscript. This work was supported by Cells were harvested at various times after the different grants from INSERM and Association pour la Recherche treatments, then fixed with 70% ethanol for at least 20 min. sur le Cancer.

References

Altin JG, Kujubu DA, Rafﬁoni S, Eveleth DD, Herschman Cortes U, Moyret-Lalle C, Falette N, Duriez C, Ghissassi F, HR and Bradshaw RA. (1991). J. Biol. Chem., 266, 5401 – Barnas C, Morel AP, Hainaut P, Magaud JP and Puisieux 5406. A. (2000). Mol. Carcinog., 27, 57 – 64. Bradbury A, Possenti R, Shooter EM and Tirone F. (1991). De Laurenzi V, Raschella G, Barcaroli D, Annicchiarico- Proc.Natl.Acad.Sci.USA,88, 3353 – 3357. Petruzzelli M, Ranalli M, Catani MV, Tanno B, Costanzo Buanne P, Corrente G, Micheli L, Palena A, Lavia P, A, Levrero M and Melino G. (2000). J. Biol. Chem., 275, Spadafora C, Lakshmana MK, Rinaldi A, Banﬁ S, Quarto 15226 – 15231. M, Bulfone A and Tirone F. (2000). Genomics, 68, 253 – el-Deiry WS, Tokino T, Velculescu VE, Levy DB, Parsons R, 263. Trent JM, Lin D, Mercer WE, Kinzler KW and Vogelstein Cimato TR, Ettinger MJ, Zhou X and Aletta JM. (1997). J. B. (1993). Cell, 75, 817 – 825. Cell Biol., 138, 1089 – 1103. Erhardt JA and Pittman RN. (1998). J. Biol. Chem., 273, Corrente G, Guardavaccaro D and Tirone F. (2002). 23517 – 23523. Neuroreport, 13, 417 – 422.

Oncogene BTG2TIS21/PC3 triggers differentiation of PC12 cells F el-Ghissassi et al 6778 Fletcher BS, Lim RW, Varnum BC, Kujubu DA, Koski RA Montagnoli A, Guardavaccaro D, Starace G and Tirone F. and Herschman HR. (1991). J. Biol. Chem., 266, 14511 – (1996). Cell Growth Differ., 7, 1327 – 1336. 14518. Rouault JP, Rimokh R, Tessa C, Paranhos G, Ffrench M, Greene LA. (1978). J. Cell Biol., 78, 747 – 755. Duret L, Garoccio M, Germain D, Samarut J and Magaud Guardavaccaro D, Corrente G, Covone F, Micheli L, JP. (1992). EMBO J., 11, 1663 – 1670. d’Agnano I, Starace G, Caruso M and Tirone F. (2000). Rouault JP, Falette N, Guehenneux F, Guillot C, Rimokh R, Mol. Cell. Biol., 20, 1797 – 1815. Wang Q, Berthet C, Moyret-Lalle C, Savatier P, Pain B, Guehenneux F, Duret L, Callanan MB, Bouhas R, Hayette Shaw P, Berger R, Samarut J, Magaud JP, Ozturk M, S, Berthet C, Samarut C, Rimokh R, Birot AM, Wang Q, Samarut C and Puisieux A. (1996). Nat. Genet., 14, 482 – Magaud JP and Rouault JP. (1997). Leukemia, 11, 370 – 486. 375. RudkinBB,LazaroviciP,LeviBZ,AbeY,FujitaKand Gupta M, England MA and Notter MF. (1987). Brain Res. Guroff G. (1989). EMBO J., 8, 3319 – 3325. Bull., 18, 555 – 561. Tirone F. (2001). J. Cell. Physiol., 187, 155 – 165. Iacopetti P, Barsacchi G, Tirone F, Maffei L and Cremisi F. Van Grunsven LA, Thomas A, Urdiales JL, Machenaud S, (1994). Mech. Dev., 47, 127 – 137. Choler P, Durand I and Rudkin BB. (1996). Oncogene, 12, IacopettiP,MicheliniM,StuckmannI,ObackB,Aaku- 855 – 862. Saraste E and Huttner WB. (1999). Proc. Natl. Acad. Sci. Vinores SA, Marangos PJ, Parma AM and Guroff G. (1981). USA, 96, 4639 – 4644. J. Neurochem., 37, 597 – 600. Ikematsu N, Yoshida Y, Kawamura-Tsuzuku J, Ohsugi M, Walden PD, Lefkowitz GK, Ficazzola M, Gitlin J and Lepor Onda M, Hirai M, Fujimoto J and Yamamoto T. (1999). H. (1998). Exp. Cell Res., 245, 19 – 26. Oncogene, 18, 7432 – 7441. Yang A, Walker N, Bronson R, Kaghad M, Oosterwegel M, Lin WJ, Gary JD, Yang MC, Clarke S and Herschman HR. Bonnin J, Vagner C, Bonnet H, Dikkes P, Sharpe A, (1996). J. Biol. Chem., 271, 15034 – 15044. McKeon F and Caput D. (2000). Nature, 404, 99 – 103. Malatesta P, Gotz M, Barsacchi G, Price J, Zoncu R and Yoshida Y, Matsuda S, Ikematsu N, Kawamura-Tsuzuku J, Cremisi F. (2000). Mech. Dev., 90, 17 – 28. Inazawa J, Umemori H and Yamamoto T. (1998). Marz P, Herget T, Lang E, Otten U and Rose-John S. (1997). Oncogene, 16, 2687 – 2693. Eur. J. Neurosci., 9, 2765 – 2773. Zhu J, Jiang J, Zhou W and Chen X. (1998). Cancer Res., 58, Matsuda S, Kawamura-Tsuzuku J, Ohsugi M, Yoshida M, 5061 – 5065. Emi M, Nakamura Y, Onda M, Yoshida Y, Nishiyama A and Yamamoto T. (1996). Oncogene, 12, 705 – 713.

Oncogene