Overexpression of the immediate early gene fra-1 inhibits proliferation, induces apoptosis, and reduces tumourigenicity of c6 glioma cells

Available online at www.sciencedirect.com R

Experimental Cell Research 291 (2003) 91–100

www.elsevier.com/locate/yexcr

Overexpression of the immediate early gene fra-1 inhibits proliferation, induces apoptosis, and reduces tumourigenicity of C6 glioma cells夞 Neelam V. Shirsat* and Sadaf A. Shaikh Neurooncology Division, Tata Memorial Centre, Advanced Centre for Treatment, Research & Education in Cancer (ACTREC), Kharghar, Navi Mumbai-410 208, India Received 27 May 2003

Abstract Hexamethylene bisacetamide (HMBA)-induced growth inhibition and differentiation of the rat C6 glioma cell line were found to be accompanied by down-regulation of the constitutively expressed fra-1 gene. In order to check if the fra-1 gene down-regulation was essential for HMBA’s growth inhibitory effect, C6 cells were stably transfected with vector expressing fra-1 cDNA under CMV promoter in either sense or antisense orientation. Contrary to the expectations, fra-1 overexpression was found to inhibit proliferation and induce morphological differentiation of C6 cells. Furthermore, all three differentiation inducers studied viz. dibutyryl cyclic AMP (dbcAMP), staurosporine, and HMBA have greater growth inhibitory effect on fra-1 overexpressing clones as compared to the parental C6 cells. dbcAMP and staurosporine not only inhibit proliferation but bring about complete apoptosis of fra-1 overexpressing clones. Spontaneous apoptosis is seen in fra-1 overexpressing clones especially in confluent cultures. fra-1 overexpression also results in substantial reduction in anchorage-independent growth and tumourigenicity of C6 cells. Overexpression of fra-1 leading to proliferation inhibition of C6 glioma cells is consistent with the concept that fra-1 functions as a negative regulator of AP-1 activity. © 2003 Elsevier Inc. All rights reserved.

Introduction Astrocytomas account for almost 90% of primary malignant brain tumours; and surgery followed by radiation therapy with or without chemotherapy is the common mode of treatment. A majority of the astrocytomas are of high grade and respond poorly to the treatment. Suicide gene therapy has also failed to improve the prognosis for astrocytoma patients. The mean life expectancy for a patient with glioblastoma multiforme is only about 1 year. It is therefore necessary to develop therapy based on the biology of these highly malignant brain tumours. To understand the molecular basis of this neoplasia, it is important to delineate the mechanism underlying proliferation and differentiation of astrocyte, the cell of origin for these tumours. dbcAMP, staurosporine, and HMBA inhibit growth and

夞 This work was supported by a grant received from the Lady Tata Memorial Trust, Mumbai. * Corresponding author. Fax: ⫹91-22-27412894. E-mail address: [email protected] (N.V. Shirsat). 0014-4827/$ – see front matter © 2003 Elsevier Inc. All rights reserved. doi:10.1016/S0014-4827(03)00346-X

induce morphological differentiation of the C6 glioma cell line, established from ENU-induced rat brain tumour [1–3]. dbcAMP, staurosporine, and HMBA act via three different signal transduction modes viz. activation of protein kinase A, and inhibition and activation of protein kinase C, respectively. We have investigated the molecular mechanism of action of these three agents on growth of C6 glioma cells. Immediate early gene c-fos is induced on treatment with all three agents. fra-1, another immediate early gene belonging to the fos family is induced by dbcAMP and staurosporine. HMBA, on the other hand, drastically reduces fra-1 mRNA levels induced on inhibition of protein synthesis by cycloheximide, both in normal and malignant astrocytes [3,4]. In this paper we will show that the effect of HMBA on fra-1 gene expression is long lasting. In C6 glioma cells, fra-1 is constitutively expressed and HMBA brings about substantial down-regulation of fra-1 gene expression. HMBA-induced terminal differentiation of murine erythroleukemia cells is known to be accompanied by induction of c-fos and down-regulation of c-myc and c-myb gene expression [5]. It has been shown that down-regulation of c-myc

92

N.V. Shirsat and S.A. Shaikh / Experimental Cell Research 291 (2003) 91–100

and c-myb is essential for the differentiation induction [6,7]. In this paper, we have investigated the possibility of fra-1 playing a role in controlling astrocyte proliferation analogous to that of c-myc or c-myb in erythroleukemia cells.

Materials and methods Materials The following chemicals and reagents were obtained from the indicated sources: Dulbecco’s modified Eagle medium (D-MEM), fetal bovine serum (FBS), G418 sulphate, and lipofectin (GIBCO-BRL, USA); dbcAMP, staurosporine, and HMBA, and [3-(4,5)-dimethylthiazol-2yl)-2,5-diphenyl-tetrazolium bromide (MTT) Luminol, (Sigma Chemical Company, USA); anti-fra-1 antibody (H-22) (Santa Cruz Biotechnology, USA): peroxidase-conjugated anti-rabbit IgG Hybond N, Hybond C, (Amersham Pharmacia Biotec, USA); [␣-32P]dCTP, Board of Radiation and Isotope Technology, (India). All other chemicals used were of the highest quality commercially available. Cell cultures C6 astrocytoma cells were obtained from the National Centre for Cell Science (India), originally from American Type Culture Collection (ATCC) (USA). Cells were maintained in D-MEM supplemented with 10% FBS. The cells were used between passages 50 and 60.

30 min. Protein was removed by precipitation with potassium acetate and DNA was obtained by precipitation with isopropanol. RNAse-treated DNA was digested with EcoRI or HindIII and electrophoretically separated on 0.8% agarose gel and blotted onto Hybond N membrane. The Southern blot was probed with (␣-32P)-dCTP-labeled fra-1 cDNA probe [9]. Genomic DNA was prepared similarly from C6 and fra-1 transfected clones grown to confluence or treated with the differentiation inducers. The DNA was separated on 1.8% agarose gel and stained with ethidium bromide for DNA ladder analysis. Cell proliferation and dose response Growth kinetics and the effect of HMBA, dbcAMP, and staurosporine on proliferation of parental C6 cell line and the fra-1 transfected clones was assessed using MTT reduction assay [10]. Cells were seeded at a density of 2500 cells/100 ␮l medium/well. Agent was added 24 h after seeding. On the third day, medium was replenished along with the agent. MTT was dissolved in phosphate-buffered saline (PBS) at a concentration of 5 mg/ml. Twenty microliters of this MTT solution was added to each well at the end of the incubation period. One hundred microliters of 10% SDS in 0.01 M HCl was added to each well to dissolve the dark blue formazan crystals. Optical density was read on an ELISA reader at the wavelength of 540 nm and a reference wavelength of 690 nm. Western blot analysis

Plasmid and transfection Rat fra-1 cDNA insert [8] in pSP65 vector was excised by digesting with EcoRI and was blunt ended using Klenow enzyme. This insert was then subcloned into the bluntended HindIII site of pRcCMV vector (Invitrogen, The Netherlands). Colonies were checked for the presence of fra-1 insert and its orientation, first by restriction digestion and then by DNA sequencing. C6 cells were transfected with the pRcCMV constructs containing fra-1 cDNA in either sense or antisense orientation with respect to CMV promoter. The transfection was done using Lipofectin reagent according to the manufacturer’s protocol. One to two micrograms of plasmid DNA mixed with 10 ␮l of Lipofectin reagent was added to 5 ⫻ 105 C6 cells plated onto 55 mm tissue culture plate. Cells were incubated with DNA for 6 h; and 24 h later were split 1:5 and selected for stably transfected clones in the presence of 800 ␮g/ml G418 sulphate. DNA analysis Genomic DNA was extracted from C6 cells and the fra-1 transfected clones in buffer containing 0.1 M Tris-HCl, pH 9.0, 0.1 M EDTA, and 1% SDS, were incubated at 70°C for

Cells were seeded, 5 ⫻ 105 per 55-mm tissue culture plate and treated with 1 mM dbcAMP, 10 nM staurosporine, or 5 mM HMBA for 3 and 6 days. Two hundred micrograms of total protein per lane was separated on 12.5% SDSPAGE, and electrophoretically transferred to nitrocellulose membrane. The Western blot was probed with anti-fra-1 antibody at a concentration of 0.5 ␮g/ml. The blot was incubated with peroxidase-conjugated anti-rabbit IgG. The bound antibody was detected by chemiluminescence. fra-1 protein levels in transfected clones were assessed similarly by Western blot analysis. Gel shift analysis Protein was extracted from parental C6 cells and the fra-1 transfected clones in cell lysis buffer (20 mM Tris [pH 7.5], 150 mM NaCl, 1 mM EDTA, 1 mM EGTA, 1% Triton X-100, 2.5 mM sodium pyrophosphate, 1 mM ␤-glycerophosphate, 1 mM sodium orthovanadate, 1 ␮g/ml leupeptin, and 1 mM PMSF). Double-stranded oligonucleotide containing the consensus AP-1 binding motif (5-⬘CGC TTG ATG ACT CAG CCG GAA-3⬘) and a mutant oligonucleotide containing “CA” to “TG” substitution in the AP-1 binding motif (Santa Cruz Biotec, USA) were labelled with

N.V. Shirsat and S.A. Shaikh / Experimental Cell Research 291 (2003) 91–100

93

Results

Fig. 1. Western blot analysis of fra-1 protein levels in C6 cells untreated or treated with 1 mM dbcAMP, 10 nM staurosporine, or 5 mM HMBA for 3 and 6 days.

[␥-32P]ATP using T4 polynucleotide kinase (Amersham, USA) and purified on a sephadex G-25 spin column. Protein (12.5 ␮g) was incubated with the labelled oligo (50,000 cpm) in 20 ␮l reaction volume in the binding buffer (5X Stock: 50 mM Tris-HCl, pH 7.5, 20% glycerol, 5 mM MgCl2, 2.5 mM EDTA, 2.5 mM DTT, 250 mM NaCl). One microgram of a nonspecific oligo (5⬘-AGC ACT CTC CAG CCT CTC ACC GCA-3⬘) was added to this reaction mixture to inhibit nonspecific binding. The mixture was incubated on ice for 30 min. The samples were loaded onto a 12 ⫻ 14 cm nondenaturing 4% acrylamide gel, which was prerun for 30 min and electrophoresed at 125 V. For Supershift analysis, prior to the addition of the labeled primer, 2 ␮l of the Supershift reagent (Santa Cruz) was added to the reaction mixture and incubated at room temperature for 1 h. Colony forming efficiency in soft agar Colony formation in soft agar was assayed as described by Freshney [11]. Briefly, 2000 cells were suspended in D-MEM with 10% FBS, 0.3% agar. These cells were plated over a 0.5 ml precast 1% agar in D-MEM in a 35 mm petri dish. After 6 – 8 days, growth in a humidified 5% CO2 incubator, colonies consisting of at least 20 –30 cells were scored under a phase contrast microscope. Simultaneously, 500 cells were plated in a 55 mm tissue culture dish in D-MEM with 10% FBS for checking the plating efficiency. After about 4 –5 days incubation, the colonies formed were fixed in chilled methanol and stained with Giemsa. Tumourigenicity of C6 glioma cells and its fra-1 transfected clones Tumourigenicity was tested by subcutaneous injection of 1 ⫻ 106 cells into the flanks of syngenic Sprague-Dawley rat pups. Four- to five-day-old rat pups were used for the study. Each pup was injected with parental C6 cells on one flank and the overexpressing fra-1 clone (TwG or TwA2) cells on the other flank. Tumours were measured using a Vernier caliper and tumour volume was calculated using the formula: volume ⫽ length ⫻ width2 ⫼ 2.

C6 cells were treated with 1 mM dbcAMP, 10 nM staurosporine, or 5 mM HMBA for a period of 3 and 6 days. Fig. 1 shows Western blot analysis of fra-1 protein levels in C6 cells treated with these agents. No significant change in fra-1 protein level is seen in dbcAMP-treated or staurosporine-treated cells. There is considerable reduction in fra-1 protein levels in C6 cells treated with HMBA. In order to check if changes in fra-1 gene expression on HMBA treatment are simply coincidental or necessary for growth inhibition, C6 cells were transfected with rat fra-1 cDNA expressed under CMV promoter in either sense or antisense orientation. About 40 clones were selected in the presence of G418 sulphate for each of the two constructs. These clones were screened for fra-1 expression by immunocytochemical analysis (data not shown). Clones TwG and TwA2 were chosen as high fra-1 expressers for the sense construct. Clones 1.2F and 1.2A2 were selected having the lowest fra-1 expression for antisense construct. fra-1 protein levels in these constructs were confirmed by Western blot analysis (Fig. 2). To demonstrate the DNA-binding activity of the overexpressed fra-1, gel shift analysis was done. A prominent broad shifted band is seen in C6 and all the fra-1 transfected clones corresponding to the AP-1 complex (Fig. 3). The intensity of this band is considerably higher in the overexpressing clones while it is lower in C6 and the antisense fra-1 clones, indicating increased AP-1 DNA binding activity in fra-1 overexpressing clones. Supershift analysis shows that in overexpressing clones the AP-1 DNA binding activity consists of c-jun and fra-1 while in C6 cells it contains predominantly c-fos and c-jun (Fig. 4), indicating that the overexpressed fra-1 is functionally active and is responsible for increased AP-1 DNA binding activity. Southern blot of genomic DNA from the fra-1 transfected clones was hybridised to the fra-1 cDNA probe (Fig. 5). Bands in addition to the one belonging to the endogenous fra-1 gene are seen in transfected clones and these are unique for each clone. This confirms the presence of exogenous fra-1 in transfected clones and the independent nature of the clones. fra-1 transfected clones show morphology distinct from parental C6 cells. C6 cells are small cuboidal cells, which vary in size and shape to some extent depending upon the cell density (Fig. 6A and 6B). 1.2F clone cells are smaller in size and show thin processes and phase-translucent appearance, especially in confluent culture (Fig. 6C and 6D).

Fig. 2. Western blot analysis of fra-1 gene expression in C6 and its transfected clones.

94

N.V. Shirsat and S.A. Shaikh / Experimental Cell Research 291 (2003) 91–100

Fig. 3. Gel shift analysis for AP-1 DNA binding activity in C6 and its fra-1 transfected clones. Lane A contains the labelled AP-1 oligomer alone. Lane B denotes gel shift analysis of TwG protein extract done in the presence of excess cold AP-1 oligomer. Lane C depicts the gel shift analysis of TwG extract done with mutant AP-1 labeled oligomer.

1.2A2 cells are intermediate in size and appearance between that of C6 cells and 1.2F cells (data not shown). Cells of the fra-1 overexpressing clones TwA2 and TwG, on the other hand, are much larger in size and flattened in appearance (Fig. 6E). In confluent cultures these clones show aggregates of rounded dead cells (Fig. 6F). About 8%–10% and 10 –15% dead cells are seen in confluent cultures of TwA2 and TwG, respectively. Fig. 7 shows growth kinetics of C6 and the fra-1 transfected clones. Cell number was calculated based on the optical density measurements in MTT assay. Clones TwG and TwA2 show significantly slower growth as compared to the parental C6 cells. The overall doubling time for TwA2 and TwG is about 22–26 h while that for C6 is 15–18 h. 1.2F and 1.2A2 growth kinetics is similar to that of parental C6 cells. The effect of HMBA, dbcAMP, and staurosporine on the proliferation of C6 cells was studied by MTT assay. HMBA in the concentration range of 2.5 to 10 mM inhibits growth of C6 cells by 22.93 ⫾ 0.63% to 91.6 ⫾ 0.36%, respectively, by the third day (Fig. 8A). About 50% reduction in growth inhibitory effect of HMBA is seen for 1.2F and 1.2A2 clones. HMBA has much higher inhibitory effect on TwG and TwA2 clones. By the sixth day, HMBA’s growth inhibitory effect on C6 cells decreases at all the concentrations. On day 6, HMBA at 5 mM concentration inhibits

Fig. 4. Supershift analysis done on C6 and TwG cell extract using labelled AP-1 primer with antibodies specific for c-fos, c-jun, and c-fra-1.

Fig. 5. Genomic Southern analysis of genomic DNA of C6 and the fra-1 transfected clones. The DNA was digested with EcoRI restriction enzyme and probed with fra-1 cDNA probe.

N.V. Shirsat and S.A. Shaikh / Experimental Cell Research 291 (2003) 91–100

Fig. 6. Phase contrast micrographs of C6 cells, clone 1.2F, and clone TwG in low density cultures (A, C, and E, respectively) and in confluent cultures (B, D, and F, respectively). Original magnification, ⫻75.

proliferation of C6 cells by only 19.59 ⫾ 0.51%. Growth of TwA2 and TwG clones on the other hand is inhibited by 60.06 ⫾ 2.5% and 91.15 ⫾ 1.11%, respectively (Fig. 8B). dbcAMP in the concentration range of 0.25 to 1.0 mM inhibits proliferation of C6 cells by 34.74 ⫾ 2.94% to 64.82 ⫾ 1.86%, respectively by third day (Fig. 9). At 0.5 mM dose, growth inhibition of C6 cells is 48.9 ⫾ 1.8%, while that for clones 1.2F and 1.2A2 is 20.83 ⫾ 3.6% and 24.57 ⫾ 1.89%, respectively. Thus growth inhibitory effect of dbcAMP on these clones is lower by about 50% as compared to that on parental C6 cells. At a concentration as low as 0.25 mM, TwA2 and TwG clones on the other hand are inhibited by as much as 60%. By day 6 of dbcAMP treatment, the majority of the cells are dead in TwG and TwA2 cultures (Fig. 11B). Staurosporine inhibits C6 cell growth by 54.8 ⫾ 3.6% to 69.33 ⫾ 2.63% in the concentration range of 2.5 to 10 nM by the third day (Fig. 10). Growth inhibition to a similar extent was observed for clones 1.2F and 1.2A2. Proliferation of TwG and TwA2 clones was inhibited by as much as

95

80%–100% on treatment with staurosporine in the range of 2.5–10 nM. On treatment with dbcAMP, staurosporine, or HMBA, C6 cells show morphological differentiation (Fig. 11A, C, and E). HMBA treatment results in an increase in cell size and broader cell shape both in parental C6 cells and in the fra-1 overexpressing clones like TwG (Fig. 11E and F). dbcAMP and staurosporine treatment of C6 cells results in thinner elongated cells (Fig. 11A and C). In TwG and TwA2 clones following this morphological change, almost all cells die by days 3– 6 of the treatment (Fig. 11B and D) Cell death seen in confluent cultures of TwG and TwA2 as well as that seen on treatment of these clones with dbcAMP or staurosporine was found to be apoptotic cell death as seen by ethidium bromide staining (data not shown). The apoptotic death was confirmed by DNA ladder analysis (Fig. 12). No cell death is seen in HMBA-treated TwG or TwA2 cells. To study if fra-1 gene expression affects anchorageindependent growth, fra-1 transfected clones were checked for the colony-forming efficiency in soft agar; C6 cells form large colonies consist of at least 70 – 80 cells by 6 – 8 days (Fig. 14C). Clones 1.2F and 1.2A2 show substantial increase in number of colonies as compared to the parental C6 cells (Fig. 13). TwG and TwA2 clones form smaller sized colonies, which appear as lumps of cells fused together. TwG and TwA2 form colonies, which are about 1/3 the size of C6 colonies and individual cell boundaries are not visible (Fig. 14D). Each of the fra-1 transfected clones was also checked for plating efficiency simultaneously. C6 and all fra-1 transfected clones show a plating efficiency of about 40%– 45% (data not shown). TwG and TwA2 colonies in plating efficiency assay are compact (Fig. 14B), while those of C6 and 1.2F and 1.2A2 colonies consist mostly of widely spaced cells (Fig. 14A), indicating lower motility of the fra-1 overexpressing cells.

Fig. 7. Growth kinetics of C6 and the fra-1 transfected clones done by MTT assay. Left axis depicts cell number calculated from the optical density measurements.

96

N.V. Shirsat and S.A. Shaikh / Experimental Cell Research 291 (2003) 91–100

Fig. 8. MTT reduction analysis of the effect of HMBA on C6 cells and the transfected clones. Y axis depicts growth inhibition expressed as percentage of control. Parts A and B show the analysis done on third and sixth days respectively. All the data points are presented as mean ⫾ standard error (vertical bars).

HMBA-induced growth inhibition of C6 cells was found to be accompanied by significant down-regulation of fra-1 gene

expression. Overexpression of fra-1 in C6 cells, however, has a growth inhibitory effect. Furthermore, HMBA inhibits proliferation of fra-1 overexpressing clones to a much greater extent as compared to that of parental C6 cells. Consistent with this observation, clones expressing fra-1 antisense construct are inhibited to a much less extent by HMBA. fra-1 gene down-regulation by HMBA therefore appears not to contribute to its growth inhibitory effect but may in fact work against the growth inhibition. While HMBA induces terminal differentiation of murine erythroleukemia cells, it fails to do so in other malignant cell types [12,13]. In C6 glioma cells as well, the growth inhibitory effect is not sustainable. HMBA-induced fra-1 gene down-regulation may be at least partially responsible for its failure to bring about sustained growth inhibition of C6 glioma cells.

Fig. 9. MTT reduction analysis of the effect of dbcAMP on C6 cells and the fra-1 transfected clones. Y axis depicts growth inhibition expressed as percentage of control. All the data points are presented as mean ⫾ standard error (vertical bars).

Fig. 10. MTT reduction analysis of the effect of staurosporine on C6 cells and the fra-1 transfected clones. Y axis depicts growth inhibition expressed as percentage of control. All the data points are presented as mean ⫾ standard error (vertical bars).

Tumourigenicity of fra-1 overexpressing clones was checked by injecting 1 ⫻ 106 cells into the flanks of 0 –2day-old syngenic rat pups. As a control, C6 cells were injected into the opposite flank of each rat pup. Tumours of measurable size were obtained by 10 –12 days after injection. In the case of TwG cells, tumours did not appear until almost 16 –20 days. About 60% reduction in tumour volume was observed for TwA2 cells as compared to the parental C6 cells (Fig. 15). For TwG cells, the tumour volume reduction was in the range of 90%–95% of the control. Discussion

N.V. Shirsat and S.A. Shaikh / Experimental Cell Research 291 (2003) 91–100

97

Fig. 12. Photograph of the ethidium bromide stained genomic DNA prepared from TwA2 and TwG cells grown to confluency (control) or treated with 5 mM HMBA, 10 nM staurosporine, or 1 mM dbcAMP and separated on 1.8% agarose gel.

Fig. 11. Phase contrast micrographs of C6 cells in culture treated with 1 mM dbcAMP (A), 10 nM staurosporine (C), or 5 mM HMBA (E). B, D, and F depict photographs of TwG cells in culture treated similarly with dbcAMP, staurosporine, or HMBA, respectively. Original magnification, ⫻75.

Overexpression of fra-1 by itself has a growth inhibitory effect. Although antisense fra-1 expressing clones do not exhibit higher proliferation rate, these clones show significant increase in their ability to grow in an anchorageindependent manner as is evident from the substantial increase in number of soft agar colonies for clones 1.2F and 1.2A2. Overexpression of fra-1 not only increases growth inhibitory effect of HMBA but also that of dbcAMP and staurosporine. Antisense fra-1 clones show less growth inhibition on treatment with dbcAMP or HMBA as compared to parental C6 cells. It is not clear why antisense clones do not exhibit any reduction in growth inhibition by staurosporine. Fra-1, a fos-related transcription factor is a component of the dimeric AP-1 transcription factor. AP-1 dimer consists of the two protein families, fos and jun. In mammalian cells, four members of the fos family (c-fos, fosB, fra-1, and fra-2) and three members of the jun family (c-jun, junB, and junD) have been identified. These proteins form jun–jun ho-

modimers or more stable fos–jun heterodimers, which activate transcription from the TPA-responsive element (TRE) containing enhancers. Both fos and jun protein families are required for cell cycle progression [14] and have been shown to induce oncogenic transformation [15]. Unlike c-fos, fra-1 lacks c-terminal transactivation domain and has therefore been proposed to act as negative regulator of AP-1 activity [16]. Consistent with this concept, unlike c-fos, fra-1 overexpression fails to transform 208 F fibroblasts [17]. An inverse relationship between

Fig. 13. Soft agar colony assay results showing the effect of fra-1 gene expression on the anchorage-independent growth. Left axis depicts number of colonies for each of the clones. All the data points are expressed as mean ⫾ standard error.

98

N.V. Shirsat and S.A. Shaikh / Experimental Cell Research 291 (2003) 91–100

Fig. 14. Photographs of Giemsa-stained colonies of C6 (A) and TwG (B) cells formed on plating the cells in tissue culture plates. Phase contrast micrographs of soft agar colonies of C6 cells (C) and those of TwG cells (D).

fra-1 and tumorigenic phenotype has been shown in cervical carcinoma cells [18]. Transgene expressing fra-1 rescues the osteopetrosis of the c-fos mutant mice in vivo by stimulating osteoclast differentiation [19]. In fra-1 transgenic mice, overexpression of fra-1 enhances osteoblast differentiation leading to increased bone formation [20]. Overexpression of fra-1 leading to proliferation inhibition of C6 glioma cells is thus consistent with these observations in other cell types in which fra-1 functions as a negative regulator of AP-1 activity. In a number of cases, however, fra-1 does not seem to act as a negative regulator of AP-1 activity. For example, fra-1 protein becomes a predominant AP-1 component upon rasinduced transformation of NIH3T3 cells [21]. Constitutive expression of fra-1 in PC12 cells results in inhibition of nerve growth factor-induced differentiation [22]. fra-1 has also been shown to induce morphological transformation and increase in vitro invasiveness and motility of epitheloid adenocarcinoma cells [23]. Overexpressing TwG and TwA2 clones, on the other hand, show reduction in anchorageindependent growth and motility as well as tumourigenicity of the cells. The activities of fra-1 may thus be cell-type specific or may depend upon its dimerisation partners. Overexpression of fra-1 in C6 cells not only inhibits growth but also induces apoptosis. Clones TwG and TwA2 show about 10%–20% spontaneous cell death. dbcAMP and staurosporine in the concentration range of 0.25–1.0 mM and 2.5–10 nM, respectively, only inhibit proliferation of

C6 cells and this inhibitory effect is not sustainable. In fra-1 overexpressing clones on the other hand, these two agents, following initial growth inhibition, induce almost 100% cell death by the fifth or sixth day. In one other case, fra-1 has been shown to induce apoptosis. fra-1 has been reported to substitute for c-fos in AP-1-mediated signal transduction in retinal apoptosis [24]. fra-1 has recently been identified as an ionizing radiation-responsive gene in the human leukemic cell line [25]. The stress-activated protein kinase p38/ SAPK2 has been found to induce fra-1 gene expression along with c-jun and GADD45 in jurkat T cells [26]. It is not known whether the stress-induced fra-1 induces or protects against apoptosis in these cell types. During vinblastine-induced growth arrest and apoptosis, fra-1 expression was found to increase and contribute to the AP-1 activity in KB-3 human carcinoma cells [27]. fra-1 therefore appears to be responsible for induction of apoptosis in other cell types as well. HMBA-treated cultures of TwG or TwA2 clones do not show the presence of dead cells. Because HMBA brings about down-regulation of fra-1 gene expression, it may be able to avert the apoptotic induction brought about by fra-1 overexpression. HMBA brings down the levels of fra-1 protein even in the overexpressing clones (data not shown). HMBA also brings about reduction in c-jun protein levels, while there is no change in c-jun levels in staurosporine or dbcAMP-treated C6 cells (data not shown). fra-1 may therefore bring about apoptotic induction in cooperation with c-jun. In a study done on human astrocytic tumours, strong fos expression was observed in 78% of grade IV and 50% of grade III astrocytomas [28]. About 50% astrocytomas of both the grades showed c-jun expression. Expression of the paired box gene pax5 was observed in these tumours in the

Fig. 15. Tumourigenicity of the fra-1 transfected clones TwA2 and TwG and that of the parental C6 cells. Y axis denotes tumour volume as measured on the day indicated on the X axis. All the data points are presented as mean ⫾ standard error (vertical bars).

N.V. Shirsat and S.A. Shaikh / Experimental Cell Research 291 (2003) 91–100

areas which expressed fos or jun and was found to correlate with increasing malignancy. Activation of AP-1 transcription factor is likely to contribute to the pathogenesis of human astrocytomas. In another recent study fra-1 has been found to be up-regulated along with c-jun in astrocytoma cell lines [29]. fra-1 up-regulation therefore is not restricted to C6 glioma cell line but seems to be true for most glioma cell lines. In this study [29], fra-1 has been shown to increase VEGF-D expression in breast cancer cell lines and has been implicated to have a similar role in the elevation of VEGF-D in glioma cell lines. However, the direct role of fra-1 in glioma cell lines has not been demonstrated. Overexpression of fra-1, if found to have an effect similar to that on the C6 glioma cell line on the proliferation and tumourigenicity of human astrocytoma cell lines, has the potential of being therapeutically useful. Recently, fra-1 has been shown to acquire transactivation potential on phosphorylation of its transactivation domain [30]. fra-1 therefore may not just act by antagonising activities of fos or jun but may also regulate specific gene expression. Members of the fos– jun family are known to interact with the transcription factors belonging to the ATF/CREB family as well as the bHLHZip USF family [31,32]. Identification of genes targeted by fra-1 during growth inhibition and apoptosis of C6 glioma cells and identification of its dimerisation and interaction partners are required to understand the molecular basis of its activities.

Acknowledgments We acknowledge Mr. Anant Sawant and Mr. Umesh Kadam for excellent technical assistance.

References [1] D.L. Davies, A. Vernadakis, Responses in astrocytic C6 glioma cells to ethanol and dibutyryl cyclic AMP, Brain Res. 389 (1986) 253–260. [2] W.T. Couldwell, J.P. Antel, M.L. Apuzzo, V.W. Yong, Inhibition of growth of established human glioma cell lines by modulators of the protein kinase-C system, J. Neurosurg. 73 (1990) 594 – 600. [3] N.V. Shirsat, J.J. Kayal, S. Shaikh, A. Mehta, Growth inhibition and differentiation of c6 glioma cells on treatment with HMBA, Cell Biol. Int. 25 (2001) 621– 627. [4] N.V. Shirsat, HMBA inhibits growth and induces differentiation of astrocytes from neonatal rat brain, Neuroreport 10 (1999) 3755–3758. [5] R.G. Ramsay, K. Ikeda, R.A. Rifkind, P.A. Marks, Changes in gene expression associated with induced differentiation of erythroleukemia: protooncogenes, globin genes, and cell division, Proc. Natl. Acad. Sci. USA 83 (1986) 6849 – 6853. [6] D. Mcclinton, J. Stafford, L. Brents, T.P. Bender, W.M. Kuehl, Differentiation of mouse erythroleukemia cells is blocked by late up-regulation of a c-myb transgene, Mol. Cell. Biol. 10 (1990) 705– 710. [7] C.M. Cultraro, T. Bino, S. Segal, Function of the c-Myc antagonist Mad1 during a molecular switch from proliferation to differentiation, Mol. Cell. Biol. 17 (1997) 2353–2359.

99

[8] D.R. Cohen, T. Curran, fra-1: a serum-inducible, cellular immediateearly gene that encodes a fos-related antigen, Mol. Cell. Biol. 8 (1988) 2063–2069. [9] A.P. Feinberg, B. Vogelstein, A technique for radiolabeling DNA restriction endonuclease fragments to high specific activity, Anal. Biochem. 132 (1983) 6 –13. [10] T. Mosmann, Rapid colorimetric assay for cellular growth and survival: application to proliferation and cytotoxic assays, J. Immunol. Methods 65 (1983) 55– 63. [11] R.I. Freshney, Culture of Animal Cells: A Manual of Basic Technique1983, Alan R Liss, New York. [12] Y. Kloog, J. Axelrod, I. Spector, Protein carboxyl methylation increases in parallel with differentiation of neuroblastoma cells, J. Neurochem. 40 (1983) 522–529. [13] W.C. Speers, C.R. Birdwell, F.J. Dixon, Chemically induced bidirectional differentiation of embryonal carcinoma cells in vitro, Am. J. Pathol. 97 (1979) 563–584. [14] K. Kovary, R. Bravo, The jun and fos protein families are both required for cell cycle progression in fibroblasts, Mol. Cell. Biol. 11 (1991) 4466 – 4472. [15] P. Angel, M. Karin, The role of Jun, Fos and the AP-1 complex in cell-proliferation and transformation, Biochim. Biophys. Acta 1072 (1991) 129 –157. [16] E. Tulchinsky, Fos family members: regulation, structure and role in oncogenic transformation, Histol. Histopathol. 15 (2000) 921–928. [17] R. Wisdon, I.M. Verma, Transformation by Fos proteins requires a c-terminal transactivation domain, Mol. Cell. Biol. 13 (1993) 7429 – 7438. [18] U. Soto, C. Denk, P. Finzer, K.J. Hutter, H. Zur Hausen, F. Rosl, Genetic complementation to non-tumorigenicity in cervical-carcinoma cells correlates with alterations in AP-1 composition, Int. J. Cancer 86 (2000) 811– 817. [19] K. Matsuo, J.M. Owens, M. Tonko, C. Elliott, T.J. Chambers, E.F. Wagner, Fosl1 is a transcriptional target of c-Fos during osteoclast differentiation, Nat. Genet. 24 (2000) 184 –187. [20] W. Jochum, J.P. David, C. Elliott, A. Wutz, H. Plenk Jr., K. Matsuo, E.F. Wagner, Increased bone formation and osteosclerosis in mice over-expressing the transcription factor Fra-1, Nat. Med. 6 (2000) 980 –984. [21] F. Mechta, D. Lallemand, C.M. Pfarr, M. Yaniv, Transformation by ras modifies AP1 composition and activity, Oncogene 14 (1997) 837– 847. [22] E. Ito, L.A. Sweterlitsch, P.B. Tran, F.J. Rauscher 3rd, R. Narayanan, Inhibition of PC-12 cell differentiation by the immediate early gene fra-1, Oncogene 5 (1990) 1755–1760. [23] O. Kustikova, D. Kramerov, M. Grigorian, V. Berezin, E. Bock, E. Lukanidin, E. Tulchinsky, Fra-1 induces morphological transformation and increases in vitro invasiveness and motility of epithelioid adenocarcinoma cells, Mol. Cell Biol. 18 (1998) 7095–7105. [24] A. Wenzel, H.P. Iseli, A. Fleischmann, F. Hafezi, C. Grimm, E.F. Wagner, C.E. Reme, Fra-1 substitutes for c-Fos in AP-1-mediated signal transduction in retinal apoptosis, J. Neurochem. 80 (2002) 1089 –1094. [25] S.A. Amundson, M. Bittner, Y. Chen, J. Trent, P. Meltzer, A.J. Fornace Jr., Fluorescent cDNA microarray hybridization reveals complexity and heterogeneity of cellular genotoxic stress responses, Oncogene 18 (1999) 3666 –3672. [26] M. Rolli-Derkinderen, M. Gaestel, p38/SAPK2-dependent gene expression in Jurkat T cells, Biol Chem 381 (2000) 193–198. [27] A. Berry, M. Goodwin, C.L. Moran, T.C. Chambers, AP-1 activation and altered AP-1 composition in association with increased phosphorylation and expression of specific Jun and Fos family proteins induced by vinblastine in KB-3 cells, Biochem. Pharmacol. 62 (2001) 581–591.

100

N.V. Shirsat and S.A. Shaikh / Experimental Cell Research 291 (2003) 91–100

[28] E.T. Stuart, C. Kioussi, A. Aguzzi, P. Gruss, Pax5 expression correlates with increasing malignancy in human astrocytomas, Clin. Cancer Res. 2 (1995) 207–214. [29] W. Debinski, B. Slagle-Webb, M.G. Achen, S.A. Stacker, E. Tulchinsky, G.Y. Gillespie, D.M. Gibo, VEGF-D is an X-linked/AP-1 regulated putative oncoangiogen in human glioblastoma multiforme, Mol. Med. 7 (2002) 598 – 608. [30] M.R. Young, R. Nair, N. Bucheimer, P. Tulsian, N. Brown, C. Chapp, T-C. Hsu, N.H. Colburn, Transactivation of Fra-1 and consequent

activation of AP-1 occur extracellular signal-regulated kinase dependently, Mol. Cell. Biol. 22 (2002) 587–598. [31] T. Hai, T. Curran, Cross-family dimerization of transcription factors Fos/Jun and ATF/CREB alters DNA binding specificity, Proc. Natl. Acad. Sci. USA 88 (1991) 3720 –3724. [32] P. Pognonec, K.E. Boulukos, C. Aperlo, M. Fujimoto, H. Ariga, A. Nomoto, H. Kato, Cross-family interaction between the bHLHZip USF and bZip Fra1 proteins results in down-regulation of AP1 activity, Oncogene 14 (1997) 2091–2098.