¹CEA, Service de Neurovirologie DSV/DRM, CRSSA, IPSC, BP 6, 92 265 Fontenay-aux-Roses cedex, France

²CEA, Laboratoire de RadioPathologie DSV/DRR, IPSC, BP 6, 92 265 Fontenay-aux-Roses cedex, France

Correspondence to: F D Boussin, CEA, Laboratoire de RadioPathologie DSV/DRR, IPSC, BP 6, 92 265 Fontenay-aux-Roses cedex, France

^aS Haik and LR Gauthier contributed equally to this work

During brain development, neuronal and glial cells are generated from neural precursors on a precise schedule involving steps of proliferation, fate commitment and differentiation. We report that telomerase activity is highly expressed during embryonic murine cortical neurogenesis and early steps of gliogenesis and progressively decreases thereafter during cortex maturation to be undetectable in the normal adult brain. We evidenced neural precursor cells (NPC) as the principal telomerase-expressing cells in primary cultures from E15 mouse embryo cortices. Their differentiation either in neurons or in glial cells lead to a down regulation of telomerase activity that was directly correlated to the decrease of telomerase core protein (mTERT) mRNA synthesis. Furthermore, we show that FGF2 (fibroblast growth factor 2), one of the main regulators of CNS development, induces a dose-dependant increase of both the proliferation of NPC and telomerase activity in primary cortical cultures without affecting the mTERT mRNA synthesis compared to that of glyceraldehyde-3-phosphate dehydrogenase (mGAPDH). Finally, we evidenced that AZT (3'-azido-2',3'-dideoxythymidine), known to inhibit telomerase activity, blocks in a dose dependant manner the FGF2-induced proliferation of NPC. Altogether, our results are in favor of an important role of telomerase activity during brain organogenesis. Oncogene (2000) 19, 2957-2966

telomerase; neural precursor cells; fibroblast growth factor 2

NPC, neural precursor cells; FGF2, fibroblast growth factor 2; NGF, nerve growth factor; TERT, telomerase reverse transcriptase protein; TER, telomerase RNA; AZT, 3'-azido-2',3'-dideoxythymidine

Introduction

Telomeres are nucleoprotein structures located at the ends of eukaryotic chromosomes (Blackburn, 1991). Mammalian telomeres consist of several kilobases of tandem TTAGGG repeats bound by specific proteins (Chong et al., 1995). They play numerous roles in the maintenance of chromosomal integrity and cell viability (Blackburn, 1991). Telomeres are synthesized de novo by a specific multisubunit ribonucleoprotein known as telomerase (Greider and Blackburn, 1985). The telomerase core protein TERT (Meyerson et al., 1997; Nakamura et al., 1997; Nakayama et al., 1998) is a reverse transcriptase that uses an internal RNA molecule (termed mTER for the mouse) as template to add telomeric repeat sequences onto chromosome ends (Blasco et al., 1995; Feng et al., 1995; Greider, 1991). In mammalians, the majority of normal somatic cells has no detectable level of telomerase activity (Prowse and Greider, 1995) whereas telomerase is expressed in germ cells, in most tumor cell lines and in a majority of cancer (Kim et al., 1994). In cells lacking telomerase, a progressive telomere shortening generally occurs with cell division because of incomplete replication of 3' ends of linear chromosomes (Allsopp et al., 1992; Harley et al., 1990) and also because of the possible action of a still unknown exonuclease (Makorov et al., 1997). This telomere shortening occurs both with passages in culture and with organ renewal in vivo and has been proposed to result in a gradual loss of replicative capacity (Allsopp et al., 1992; Harley et al., 1990). Indeed telomere shortening may act as a mitotic clock to eventually signal cell cycle exit and cellular senescence (Harley et al., 1990, Harley, 1991). This telomere hypothesis has been further sustained by the observation that ectopic expression of telomerase in normal human cells results in an extended lifespan (Bodnar et al., 1998; Vaziri and Benchimol, 1998). Therefore, dysregulation of telomerase expression in cancer cells may play an important role in facilitating tumor development.

Telomerase is normally expressed in fetal tissues and in germ cells, and at lower levels in activated lymphocytes and cycling stem cells in adults. Its precise role in normal cells has not yet been fully explored. Telomerase maintenance of telomeres has been proposed to play a critical role in preserving genomic stability and long term viability of highly proliferative organ systems (Lee et al., 1998). Recent works have shown that after successive generation mice genetically deficient for the telomerase RNA (mTER null mice) present defects in germ cell growth, uterine and intestinal villus atrophy, impaired lymphocyte mitogenesis, diminished hematopoietic reserves, ulcerative dermatitis and alteration of nervous system development (Blasco et al., 1997; Herrera et al., 1999a,b).

To date, very few informations are available about telomerase expression and regulation in fetal tissues and its role in organogenesis remains unclear. This is particularly the case of the developing brain. Total brain of murine embryo has been shown to express significant level of telomerase activity (Greenberg et al., 1998; Martin-Rivera et al., 1998; Prowse and Greider, 1995) but precise time schedule, tissular and cellular regulation and the nature of the telomerase expressing cells are still unknown. mTER has been shown to be down-regulated in mouse brain at day 16 after birth (Blasco et al., 1995). It has been reported that mice deficient for telomerase (mTER null mice) fail to close neural tube, the penetrance of this effect increasing with the generation number (Herrera et al., 1999a). Cortical neurons and glia arise from multipotential cells with stem-like properties in embryonic cortex (Gray et al., 1988; Turner and Cepko, 1987). These neural stem cells could self renew and differentiate into cells of all neural lineages via restricted progenitor cells influenced by intrinsic and environmental factors (Cattaneo and McKay, 1990; Qian et al., 1997; Reynolds et al., 1992). FGF2 (fibroblast growth factor 2), also known as basic fibroblast growth factor is widely expressed in embryonic CNS (Giordano et al., 1992). There are increasing evidences that FGF2 may be one of the main regulators of CNS development (Dono et al., 1998; Murphy et al., 1994; Ray et al., 1997; Vaccarino et al., 1999). FGF2 has been shown to have a mitogenic action on cortical progenitor cell but also to regulate the generation of neurons and glia (Qian et al., 1997).

We therefore investigated telomerase activity during mouse brain development. We showed that telomerase expression is strictly regulated depending on brain area and that neural precursor cells (NPC) are the principal source of telomerase activity in the developing brain. Moreover, we evidenced in primary cultures derived from fetal murine brain that FGF2 dramatically up-regulated telomerase activity in NPC through a post-transcriptional mechanism. These data together with the observation that AZT (3'-azido-2',3'-dideoxythymidine), a telomerase inhibitor, blocks the FGF2-induced proliferation of cortical precursors highly suggest that telomerase play an important role during brain organogenesis.

Results

Telomerase activity during cortical and cerebellar development

We assessed telomerase activity in the developing murine brain using a Telomeric Repeat Amplification Protocol (TRAP; Kim et al., 1994) followed by ELISA detection. High levels of telomerase activity were detected in cerebral cortices of E15 C57BL/6 mouse embryos that progressively decreased during cortical maturation to become low in cortices of 16-day-old mice (Figure 1). We found no activity in adult brain, whereas telomerase was active in adult spleen, thymus and inguinal and mesenteral lymph nodes as previously reported (Prowse and Greider, 1995; Yamaguchi et al., 1998); (data not shown). However, we found that telomerase activity was largely higher in the developing brain than in adult positive tissues. High telomerase activity in E15 cortices was coincident with the peak of neurogenesis (E10-E18) and early stages of gliogenesis (Qian et al., 1997). Interestingly, telomerase activity was each time greater in cerebellum than in cortex from 8- and 16-day-old mice. We performed TRAP assays on decreasing amounts of cerebellum and cortex extracts obtained from 8-day-old mice and found similar OD values with approximately 30 time less amounts of extracts from cerebellum than from cortex (Figure 1c). Thus, telomerase activity appears directly correlated to the respective stages of development of these brain structures. Furthermore, as observed during cortex maturation in embryos, telomerase activity decreased during cerebellar maturation (Figure 1). Altogether, these data suggest that telomerase expression is strictly time regulated during brain development and depends on brain areas and neural cell differentiation.

Telomerase activity is down-regulated during neuronal differentiation of cortical precursors in vitro

We then investigated telomerase expression in primary cortical cultures from E15 mouse embryos. Twenty-four hours after seeding, cultures contained approximately 50% of neurons (MAP2-microtubule-associated protein 2-positive cells), 30% of cells that express nestin, a marker of neural precursor cells (NPC) (Lendahl et al., 1990), and no detectable GFAP-positive cell. At this time, cultures exhibited high levels of telomerase activity (Figure 2a,b). Thereafter, when cells were cultured in defined medium, they markedly decreased their telomerase activity (Figure 2a,b). In those culture conditions, rare division occurred and cells from E15 mouse embryos progressively lost nestin expression and mainly differentiated in neurons. Indeed, after 8 days, cultures contained more than 95% of neurons (MAP2-positive cells), less than 5% of GFAP-positive astrocytes and less than 1% of microglial cells (i.e. either Mac1- or isolectin-B4 positive cells). Thus, in vitro neuronal differentiation of NPC was associated to a progressive decrease of telomerase activity. Telomerase expression has been shown to be tightly linked to the cell cycle (Zhu et al., 1996). Contrary to neurons, primary astrocytes could proliferate actively in vitro. We thus tested the contribution of cycling astrocytes in the telomerase activity detected in primary cortical cultures. A similar decrease of telomerase activity was obtained when cortical cultures from E15 mouse embryos were maintained in 3% fetal calf serum-supplemented medium that led to a greater astroglial proliferation (indeed, closed to 10% of cells were GFAP-positive after 8 days in cultures) indicating that proliferating astrocytes are not the principal source of telomerase activity in primary fetal cortical cultures. This was further sustained by the lack of detection of telomerase activity in astrocyte cultures derived from cortices of one day post-natal mice (Figure 2a) suggesting that differentiated astrocytes did not express detectable telomerase activity. The progressive decrease in telomerase activity observed in primary cortical cell cultures was thus directly correlated to the differentiation of NPC, which appear to be the principal telomerase expressing cells in those cultures.

FGF 2 up-regulates telomerase activity in primary cortical cells

In order to precise the regulation of telomerase activity in NPC, we treated fetal cortical cells with growth factors and neurotrophins involved in cortical development. In murine cerebral cortex, both FGF2 mRNA and protein levels have been shown to increase dramatically from E10 to E18 and to drop to lower levels postnatally (Qian et al., 1997) similarly as that we observed for telomerase activity. Treatment of cortical cells with FGF2 immediatly after plating induced a significant increase in a dose dependent manner of ³H-thymidine incorporation (Figure 3a). As previously reported (Ghosh and Greenberg, 1995), cell proliferation induced by FGF2 was directly linked to a dose dependant increase of the percentage of nestin-positive cells (Figure 3b,c). Indeed, more than 40% of cells expressed nestin after 4 days of cultures in the presence of 40 ng/ml of FGF2, whereas they were less than 10% in untreated controls (Figure 3b,c). We observed that FGF2-induced proliferation of nestin-positive cells correlated with a dose-dependant increase of telomerase activity (Figure 4a). After 4 days of culture, an approximately eightfold increase of telomerase activity based on ELISA values was detected in extracts from cells treated with 40 ng/ml of FGF2 compared to that of untreated controls (Figure 4b). By contrast, Nerve growth factor (NGF) that did not induce nestin-positive cell proliferation, did not increase telomerase activity in primary fetal cortical cultures. The neurotrophin NT3 has been demonstrated to antagonize the proliferative effects of FGF2 and to enhance neuronal differentiation of cortical precursor cells (Ghosh and Greenberg, 1995). When primary cortical cultures were treated immediately after platting both with 40 ng/ml of FGF2 and 100 ng/ml of the neurotrophin NT3, we observed a significant reduction of telomerase activity by comparison with controls treated with FGF2 only (data not shown). These results confirmed the effect of FGF2 on NPC telomerase expression and the down-regulation of telomerase during neural differentiation. In additionnal experiments we verified that FGF2 did not induce telomerase activity in astrocyte cultures derived from cortices of 1-day-old post-natal mice (Figure 4a).

mTERT mRNA expression in primary cortical cells

To further characterize telomerase regulation in primary cortical cells, we investigated the expression of mTERT (mouse telomerase reverse transcriptase) mRNA using semi-quantitative RT-PCR. Expression of the catalytic subunit of telomerase has been shown to be tightly linked to the regulation of the enzymatic activity (Greenberg et al., 1998; Martin-Rivera et al., 1998). As previously described, we found mTERT mRNA in mouse brain tissues regardless the presence or the absence of detectable telomerase activity (Figure 5a). However, they were largely less abundant in adult cortex (telomerase negative) compared to developing brain (telomerase positive). In vitro, we observed that neuronal differentiation led to a significant reduction of mTERT mRNA synthesis in primary cortical cells from E15 mouse embryos, directly correlated with the decline in telomerase activity (Figure 2c). However, FGF2-induced up regulation of telomerase activity was not correlated to an overexpression of mTERT mRNA compared to that of GAPDH (Figure 5a,b). Indeed after 1 day, mTERT mRNA levels remained unaffected by FGF2 treatment. These results may suggest that FGF2 regulates telomerase activity post-transcriptionnally in NPC.

AZT inhibits both telomerase activity and proliferation of NPC

In order to investigate the role of telomerase in dividing cortical precursor, we studied the effect of AZT on primary cultures. AZT has been shown to inhibit telomerase activity and to cause a progressive telomere shortening in various immortalized cell lines (Gomez et al., 1998; Strahl and Blackburn, 1996). We found that AZT blocked in a dose dependent manner the up-regulation of telomerase activity induced by FGF2 in cortical cultures (Figure 6a,b). This effect was not linked to AZT toxicity : no cell death was detected after 1 and 4 days in AZT-treated cultures using MTT assay (data not show). Reduced telomerase activity was rather directly related to a reduced proliferation of NPC as a consequence of AZT treatments. Indeed we observed a dose dependent decrease of both thymidine incorporation (Figure 6c) and percentage of nestin positive cells in treated cultures (Figure 6d).

Discussion

The development of the mature central nervous system from pluripotent neural cells involves a precise balancing of primary genesis of neurons and glia, downregulation of proliferation and removal of excess cells with different topographical kinetics (Hiratochi et al., 1998; Reynolds and Weiss, 1996). Pluripotent neural cells undergo steps of proliferation, fate commitment and differentiation. Our study showed that telomerase activity is highly expressed during embryonic cortical neurogenesis and early step of gliogenesis and progressively decreases thereafter during cortex maturation to be undetectable in the normal adult brain (Blasco et al., 1995; Greenberg et al., 1998; Martin-Rivera et al., 1998; Prowse and Greider, 1995). A similar kinetic exists during the postnatal development of the cerebellum, suggesting that telomerase expression and regulation are directly related to the proliferation and differentiation of neural precursors. We further evidenced neural precursor cells (NPC) as the principal telomerase-expressing cells in primary cultures of E15 mouse embryo cortices. We found that either their neuronal differentiation when cultured in defined medium, or their glial differentiation (thus in proliferating cells) lead to a down regulation of telomerase activity, similarly as that previously reported for tumor cell line of neural origin treated with differentiating agents (Fu et al., 1999; Kruk et al., 1996). It should be noted that we found that proliferating astrocytes derived from immature brain do not express detectable telomerase activity. Thus strong evidences indicate that NPC are certainly the major source of telomerase activity in the developing brain.

FGF2 plays many important roles in the growth and morphogenesis of the cerebral cortex (Vaccarino et al., 1999) and its mitogenic effects on NPC has been widely described (Murphy et al., 1994; Qian et al., 1997; Ray et al., 1997). We have shown that FGF2 also modulates telomerase activity expressed by NPC in primary cortical cultures in a dose dependent manner. Concommitant upregulation of telomerase with FGF-2-induced NPC proliferation is consistent with the data reported for the hematopoietic system (Engelhardt et al., 1997). Telomerase has been shown to be upregulated in response to cytokine-induced proliferation and cell cycle activation in primitive resting hematopoietic cells that otherwise expressed low or undectable enzymatic activity (Weng et al., 1996), similarly as that we have observed for proliferating NPC. There is strong evidence that FGF2 effectively modulates telomerase in vivo because telomerase activity is directly correlated to the level of FGF2 expression during brain development. Mice genetically deficient for the telomerase RNA (mTR) exhibit abnormalities of neural tube formation which depends on the proliferation of multipotent precursors (Herrera et al., 1999a). FGF2 has been shown to be involved in neural tube formation by mouse neural precursor cells in vitro (Murphy et al., 1994). Taken together these data suggest that regulation by FGF2 of telomerase activity can be determinant for the brain development. Furthermore reexpression of telomerase is thought to be a significant step in oncogenesis. In human neuroblastomas and glioblastomas, detection of telomerase activity is considered as a pronostic factor (Hiyama et al., 1995). FGF2 has been reported to promote proliferation of various glioblastoma and neuroblastoma cells in vitro and in vivo in an autocrine manner (Janet et al., 1995; Joy et al., 1997). Our findings concerning primary cortical cells suggest that FGF2-upregulation of telomerase activity can be also involved in brain tumor development.

Telomerase activity has been reported to be mainly modulated at the transcriptional level by the expression of the mRNA encoding the catalytic subunit TERT in cells of various origins (Weng et al., 1997; Xu et al., 1999). Indeed, although some variations of their expression could occur (Reichman et al., 1997), mRNAs encoding the other known enzymatic subunits are nearly ubiquitously expressed in telomerase negative cells (Ito et al., 1998; Takakura et al., 1998). Down-regulation of telomerase activity has been found to be a frequent response to the induction of differentiation of immortalized cell lines and generally linked to a decrease of mRNA encoding the catalytic subunit of the enzyme, TERT (Holt et al., 1996; Kruk et al., 1996; Sharma et al., 1995; Xu et al., 1999). We also found that downregulation of telomerase activity both in vivo, in mouse cortices, and in vitro, during NPC differentiation, was directly linked to the down regulation of mTERT mRNA synthesis. But conversely, the lack of increase of mTERT mRNA expression in FGF2-treated primary cortical cultures highly suggests that FGF-2 acts on telomerase activity at a post-transcriptional level. Recently, it has been reported that Akt protein kinase (also known as PKB) enhances human telomerase activity through the phosphorylation of telomerase catalytic subunit (Kang et al., 1999). Akt is one of the down stream targets of phosphatidylinositol 3-kinase (PI3-K) that is a signaling molecule associated with receptors of growth factors (Datta et al., 1996). PI3-K/Akt pathway has been shown to be involved in the intracellular signaling of FGF2 in cells of neural origin (Miho et al., 1999; van Weering et al., 1998) and thus should be proposed to mediate the FGF-2-induced telomerase upregulation. The role of this pathway in telomerase regulation is also sustained by studies involving PTEN, an anti-oncogene that exerts its role as a tumor suppressor by negatively regulating the Akt signaling pathway (Stambolic et al., 1998). Indeed, restoration of wild-type PTEN expression leads to apoptosis, induces differentiation, and reduces telomerase activity in human glioma cells U87 MG (Tian et al., 1999) a cell line that express FGF2 (Miyagi et al., 1998).

AZT has been shown to inhibit telomerase activity and, as a consequence, to induce a reduction of telomere length in different cell types of various origins including mouse cells and finally to result in some cases in cell proliferation arrest (Melana et al., 1998; Multani et al., 1998; Strahl and Blackburn, 1996). AZT is thought to be added preferentially by telomerase at telomere ends and as a consequence to block further telomere elongation by the enzyme (Olivero and Poirier, 1993; Strahl and Blackburn, 1996). We evidenced that AZT treatment strikingly resulted in a dose dependent inhibition of FGF2-induced NPC proliferation associated with a decrease of telomerase activity. Although we cannot exclude the possibility that AZT has an effect on NPC unrelated to telomerase activity, our observation suggests that the block of telomere elongation by telomerase induces almost instantly an arrest of cell proliferation and raises the question of the precise roles of telomerase activity in NPC. A simple view could be that telomerase elongates telomeres and thus increases their cellular division capacity by the prevention of replicative senescence. However, the rapid inhibition of FGF2-induced proliferation by AZT suggests a more direct involvement of telomere maintenance in the regulation of proliferation and differentiation of NPC. This mechanism could possibly be related to the protective function of telomerase allowing cell proliferation that has been proposed by Zhu et al. (1999) to be distinguishable from net telomere lengthening in human fibroblast transfected by hTERT. This possible effect of AZT should be thus further investigated, this could be done in part by the determination of telomere lengths in NPC.

The specificity of the mammalian brain is that, except for the olfactory bulb and to a lesser extent for hippocampus, adult brain cannot generate new neurons and has reduced capacities to generate new macroglial cells (Altman and Das, 1965). The isolation, propagation, and manipulation of neural stem cells may have important therapeutic applications in neurological disorders (Cameron et al., 1998; Gage et al., 1995). Understanding the precise role and regulation of telomerase in NPC appears thus of crucial interest.

Materials and methods

Primary mouse cortical cultures

Primary mouse cortical cell cultures were established from 15-day-old C57BL6 mouse embryos. Cortices were dissected under binocular microscope, carefully freed of meninges and incubated in trypsin/EDTA solution for 10 min at 37°C. After trypsin inactivation, cells were dissociated mechanically with a flame-narrowed Pasteur pipette in DMEM (Gibco) containing 1% fetal calf serum (FCS). The suspension was pelleted by centrifugation and resuspended in DMEM high glucose containing B27 (Gibco) and 3% FCS. Twelve well plates coated with 10 g/ml of poly-D lysine were seeded at 1´10⁶ cells per well in DMEM B27 supplemented by 3% FCS. Cultures were kept in water saturated incubator with an atmosphere of 95% air 5% CO2 for 2 days. Medium was then changed for serum free DMEM-B27. After 1 or 2 weeks in culture, cells were immunocytochemically defined as 95% pure in neurons (according to MAP2 immunolabeling) and containined less than 5% glial cells (about 4-5 % GFAP positive cells and less than 1% MAC-1 or IB4 positive cells). In some experiments, cortical cultures were maintained in a 3% FCS supplemented medium. For experiments involving growth factors treatments, cortical cells were obtained using a serum free protocol: after dissociation in DMEM, cells were washed three times in DMEM and then cultured in DMEM-B27 containing 1-40 ng/ml of either recombinant mouse nerve growth factor (NGF, Sigma) or human FGF2 (Sigma). In some experiments, AZT (Glaxo-Wellcome) was added at a final concentration of 5-50 M in cultures.

Primary mouse astrocyte cultures

Astrocytes were purified from cortical cultures of neonatal C57BL/6 mice as previously described (Peyrin et al., 1999). Briefly, cortices from 1-day-old mice were dissected and carefully freed of meninges. After chemical and mechanical dissociation, cells were plated and fed twice weekly with DMEM supplemented with 10% FCS. After 13-15 days of culture, contaminating microglial cells were dislodged from cultures by shaking for 2 h at 220 r.p.m. After washing, astrocytes were trypsinized, and a 30 min. adherence step allowed to eliminate the remaining microglial cells. Cells were then cultured in 6-well plates. Seventy per cent confluent astrocyte cultures were treated with growth factors in presence of 10 or 3% of FCS or after a serum deprivation of 4 days.

Immunocytochemistry

For immunocytochemistry studies, cells were grown in Labtek^Ò cultures dishes (Nunc). Before staining, cells were rinsed twice and fixed in 4% paraformaldehyde. For intracellular labeling cells were permeabilized with a 0.1% Triton X-100 solution. Saturation was achieved by incubation in PBS supplemented with 5% goat serum and 5% FCS for 1 h at room temperature (RT). Cells were then incubated with various primary antibodies directed against MAP2 (1/100, Boehringer-Mannheim), GFAP (1/50, Boehringer-Mannheim), MAC-1 (1/100, Boehringer-Mannheim), nestin (1/50, Pharmingen) for 45 min at RT. Secondary antibodies were conjugated either with Texas-red or FITC (Cliniscience). Nuclear staining was achieved by incubation with DAPI. Slides were mounted under Fluoromount^Ò (Southern Biotechnology Associates) and were examined under a fluorescence microscope (axiovert, Zeiss).

Proliferation assays

Cortical cells obtained using a serum-free protocol were cultured in 96-well plates (100 000 cells per well) in serum free DMEM-B27. After 2 days, 37´10³ TBq of tritiated thymidine (³H-TTP, Amersham) was added to each well for 10 h. Cells were then harvested onto glassfiber filters by an automated cell harvester (Skatron) and incorporated radioactivity was measured in a scintillation counter (Microplus, EGG Wallac). Experiments were performed in sextuplicates.

MTT assay

Medium was replaced by 100 l PBS containing 0.4 mg/ml solution of 3,[4,5 dimethylthiazol-2yl]-2,5 diphenyltetrazolium bromide (MTT; Sigma). After 2 h at 37°C, the solution was removed and the produced blue formazan was solubilized with a propanol 2/HCl 1N (92 : 8) solution. The optical density was measured at 540/630 nm (Bio-tek instruments inc).

Telomerase activity assay (TRAP-ELISA)

Cortex and cerebellum of C57BL/16 mouse were dissected at various stages of development, and frozen at -80°C. Tissue fragments were homogenized in a sterile mortar and freeze by adding liquid nitrogen. Samples were then pulverized by griding with a matching pestle transferred in a sterile 1.5 ml centrifuge and resuspended in ice-cold CHAPS lysis buffer. After incubation at 4°C for 30 min and a centrifugation of 16 000 g for 25 min at 4°C, cell extracts were kept frozen at -80°C.

Cultured cells were washed three times in PBS and scraped in 2 ml of PBS, after a centrifugation, cells were resuspended in ice-cold CHAPS Iysis buffer as described for tissue fragments. Protein concentrations were determined using the Bio-Rad protein assay (Bio-Rad laboratories, CA, USA).

Telomerase activity was measured using a polymerase chain reaction-based telomeric repeat amplification protocol (Kim et al., 1994) with an ELISA detection (TRAP-ELISA, Oncor) according to manufacturer recommendations. In final, the amount of TRAP product was determined by the measure of the absorbance at 450 nm and 690 nm using a microtiter plate reader. This assay is semiquantitative, linear conditions lies between 0.2 and 1.5 of DO values, allowing the comparison of different samples in a same experiment. Therefore, in order to obtain DO values in the linear range, most of the time, telomerase activity assays were performed on 1 g of proteins from tissue fragments and 125 ng of cell cultures extracts. In each set of experiments, aliquots of all samples were treated by RNase (37°C for 20 min) in order to ascertain the specificity of the reaction. Results were confirmed by a direct visualization of the TRAP ladder by a 12.5% nondenaturing polyacrylamide gel electrophoresis (PAGE) and staining with gelstar nucleic Acid Gel Stain (FMC-Bioproducts).

Reverse-transcription/polymerase chain reaction (RT-PCR)

Total RNAs were isolated from 1-2´10⁶ cells lysed with TRIzol Reagent/Chloroform (Gibco) and treated with DNase/RNase free (5 U) and RNase inhibitor (40 U) at 37°C for 30 min. The DNase was then inactivated by an incubation at 94°C for 5 min. cDNAs were then synthesized at 42°C for 2 h from 1 g total RNA using MuLV Transcriptase Inverse (300 U, Gibco) in a 60 l reaction mixture containing oligodT 12-18 (0.4 g/l, Sigma), RNase inhibitor (48 U, Boehringer-Mannheim) and dNTP (5 mM). PCR reactions were then performed on 2 l of cDNA, except special specification, using 1.25 U of Taq DNA polymerase (Pharmacia) in 50 l of 5 mM of dNTP, and 25 M of primers in a thermal cycler (MJ-Research). mTERT mRNA were amplified by 33 cycles at 94°C for 45 s, 59°C for 2 min and 72°C for 1 min, using the primer pairs 5'-TGAGCGGACAAAACATCC-3' (mTERT-AN1) and 5'-AGGCTCGTCTTAATTGAGGT-3' (mTERT-AC1). Glyceraldehyde-3-phosphate dehydrogenase (mGAPDH) mRNAs were amplified by 23 cycles of 94°C for 45 s, 56°C for 2 min and 72°C for 1 min using the primers pairs 5'-TGAAGGTCGGTGTGAA>CGGATTGGC-3' (mGAPDH-AN1) and 5'-CATGTAGGCCATGAGGTCCACCAC-3' (mGAPDH-AC1). These conditions allow PCR amplifications in linear conditions for both fragments. PCR products were electrophoresed in 1.5% agarose gel and stained with Ethidium bromide. Densitometric analysis was performed with NIH Image 1.60.

Allsopp RC, Vaziri H, Patterson C, Goldstein S, Younglai EV, Futcher AB, Greider CW and Harley CB. (1992). Proc Natl Acad Sci USA 89, 10114-10118. MEDLINE

Altman J and Das GD. (1965). J Comp Neurol 124, 319-335. MEDLINE

Blackburn EH. (1991). Nature 350, 569-573. MEDLINE

Blasco MA, Funk W, Villeponteau B and Greider CW. (1995). Science 269, 1267-1270. MEDLINE

Blasco MA, Lee HW, Hande MP, Samper E, Lansdorp PM, DePinho RA and Greider CW. (1997). Cell 91, 25-34. Article MEDLINE

Bodnar AG, Ouellette M, Frolkis M, Holt SE, Chiu CP, Morin GB, Harley CB, Shay JW, Lichtsteiner S and Wright WE. (1998). Science 279, 349-352. Article MEDLINE

Cameron HA, Hazel TG and McKay RD. (1998). J Neurobiol 36, 287-306. Article MEDLINE

Cattaneo E and McKay R. (1990). Nature 347, 762-765. MEDLINE

Chong L, van Steensel B, Broccoli D, Erdjument-Bromage H, Hanish J, Tempst P and de Lange T. (1995). Science 270, 1663-1667. MEDLINE

Datta K, Bellacosa A, Chan TO and Tsichlis PN. (1996). J Biol Chem 271, 30835-30839. MEDLINE

Dono R, Texido G, Dussel R, Ehmke H and Zeller R. (1998). EMBO J 17, 4213-4225. Article MEDLINE

Engelhardt M, Kumar R, Albanell J, Pettengell R, Han W and Moore MA. (1997). Blood 90, 182-193. MEDLINE

Feng J, Funk WD, Wang SS, Weinrich SL, Avilion AA, Chiu CP, Adams RR, Chang E, Allsopp RC, Yu J, Le S, West ND, Harley CB, Andrews WH, Grieder CW and Villeponteau B. (1995). Science 269, 1236-1241. MEDLINE

Fu W, Begley JG, Killen MW and Mattson MP. (1999). J Biol Chem 274, 7264-7271. Article MEDLINE

Gage FH, Coatesn PW, Palmer TD, Kuhn HG, Fisher LJ, Suhonen JO, Peterson DA, Suhr ST and Ray J. (1995). Proc Natl Acad Sci USA 92, 11879-11883. MEDLINE

Ghosh A and Greenberg ME. (1995). Neuron 15, 89-103. MEDLINE

Giordano S, Sherman L, Lyman W and Morrison R. (1992). Dev Biol 152, 293-303. MEDLINE

Gomez DE, Tejera AM and Olivero OA. (1998). Biochem Biophys Res Commun 246, 107-110. MEDLINE

Gray GE, Glover JC, Majors J and Sanes JR. (1988). Proc Natl Acad Sci USA 85, 7356-7360. MEDLINE

Greenberg RA, Allsopp RC, Chin L, Morin GB and DePinho RA. (1998). Oncogene 16, 1723-1730. MEDLINE

Greider CW. (1991). Mol Cell Biol 11, 4572-4580. MEDLINE

Greider CW and Blackburn EH. (1985). Cell 43, 405-413. MEDLINE

Harley CB. (1991). Mutat Res 256, 271-282. MEDLINE

Harley CB, Futcher AB and Greider CW. (1990). Nature 345, 458-460. MEDLINE

Herrera E, Samper E and Blasco MA. (1999a). EMBO J 18, 1172-1181. Article MEDLINE

Herrera E, Samper E, Martin-Caballero J, Flores JM, Lee HW and Blasco MA. (1999b). EMBO J 18, 2950-2960. Article MEDLINE

Hiratochi M, Fujiwara A, Kitani H, Iguchi T, Sakakura T and Tomooka Y. (1998). Dev Growth Differ 40, 59-65. MEDLINE

Hiyama E, Hiyama K, Yokoyama T, Matsuura Y, Piatyszek MA and Shay JW. (1995). Nat Med 1, 249-255. MEDLINE

Holt SE, Wright WE and Shay JW. (1996). Mol Cell Biol 16, 2932-2939. MEDLINE

Ito H, Kyo S, Kanaya T, Takakura M, Inoue M and Namiki M. (1998). Clin Cancer Res 4, 1603-1608. MEDLINE

Janet T, Ludecke G, Otten U and Unsicker K. (1995). J Neurosci Res 40, 707-715. MEDLINE

Joy A, Moffett J, Neary K, Mordechai E, Stachowiak EK, Coons S, Rankin-Shapiro J, Florkiewicz RZ and Stachowiak MK. (1997). Oncogene 14, 171-183. MEDLINE

Kang SS, Kwon T, Kwon DY and Do SI. (1999). J Biol Chem 274, 13085-13090. MEDLINE

Kim NW, Piatyszek MA, Prowse KR, Harley CB, West MD, Ho PL, Coviello GM, Wright WE, Weinrich SL and Shay JW. (1994). Science 266, 2011-2015. MEDLINE

Kruk PA, Balajee AS, Rao KS and Bohr VA. (1996). Biochem Biophys Res Commun 224, 487-492. Article MEDLINE

Lee HW, Blasco MA, Gottlieb GJ, Horner JW, Greider CW and DePinho RA. (1998). Nature 392, 569-574. Article MEDLINE

Lendahl U, Zimmerman LB and McKay RD. (1990). Cell 60, 585-595. MEDLINE

Makarov VL, Hirose Y and Langmore JP. (1997). Cell 88, 657-566. Article MEDLINE

Martin-Rivera L, Herrera E, Albar JP and Blasco MA. (1998). Proc Natl Acad Sci USA 95, 10471-10476. Article MEDLINE

Melana SM, Holland JF and Pogo BG. (1998). Clin Cancer Res 4, 693-696. MEDLINE

Meyerson M, Counter CM, Eaton EN, Ellisen LW, Steiner P, Caddle SD, Ziaugra L, Beijersbergen RL, Davidoff MJ, Liu Q, Bacchetti S, Haber DA and Weinberg RA. (1997). Cell 90, 785-795. MEDLINE

Miho Y, Kouroku Y, Fujita E, Mukasa T, Urase K, Kasahara T, Isoai A, Momoi MY and Momoi T. (1999). Cell Death Differ 6, 463-470. MEDLINE

Miyagi N, Kato S, Terasaki M, Shigemori M and Morimatsu M. (1998). Int J Oncol 12, 1085-1090. MEDLINE

Multani AS, Furlong C and Pathak S. (1998). Int J Oncol 13, 923-925. MEDLINE

Murphy M, Reid K, Ford M, Furness JB and Bartlett PF. (1994). Development 120, 3519-3528. MEDLINE

Nakamura TM, Morin GB, Chapman KB, Weinrich SL, Andrews WH, Lingner J, Harley CB and Cech TR. (1997). Science 277, 955-959. Article MEDLINE

Nakayama J, Tahara H, Tahara E, Saito M, Ito K, Nakamura H, Nakanishi T, Ide T and Ishikawa F. (1998). Nat Genet 18, 65-68. MEDLINE

Olivero OA and Poirier MC. (1993). Mol Carcinog 8, 81-88. MEDLINE

Peyrin JM, Lasmezas CI, Haik S, Tagliavini F, Salmona M, Williams A, Richie D, Deslys JP and Dormont D. (1999). Neuroreport 10, 723-729. MEDLINE

Prowse KR and Greider CW. (1995). Proc Natl Acad Sci USA 92, 4818-4822. MEDLINE

Qian X, Davis AA, Goderie SK and Temple S. (1997). Neuron 18, 81-93. MEDLINE

Ray J, Baird A and Gage FH. (1997). Proc Natl Acad Sci USA 94, 7047-7052. Article MEDLINE

Reichman TW, Albanell J, Wang X, Moore MA and Studzinski GP. (1997). J Cell Biochem 67, 13-23. Article MEDLINE

Reynolds BA, Tetzlaff W and Weiss S. (1992). J Neurosci 12, 4565-4574. MEDLINE

Reynolds BA and Weiss S. (1996). Dev Biol 175, 1-13. Article MEDLINE

Sharma HW, Sokoloski JA, Perez JR, Maltese JY, Sartorelli AC, Stein CA, Nichols G, Khaled Z, Telang NT and Narayanan R. (1995). Proc Natl Acad Sci USA 92, 12343-12346. MEDLINE

Stambolic V, Suzuki A, de la Pompa JL, Brothers GM, Mirtsos C, Sasaki T, Ruland J, Penninger JM, Siderovski DP and Mak TW. (1998). Cell 95, 29-39. MEDLINE

Strahl C and Blackburn EH. (1996). Mol Cell Biol 16, 53-65. MEDLINE

Takakura M, Kyo S, Kanaya T, Tanaka M and Inoue M. (1998). Cancer Res 58, 1558-1561. MEDLINE

Tian XX, Pang JC, To SS and Ng HK. (1999). J Neuropathol Exp Neurol 58, 472-479. MEDLINE

Turner DL and Cepko CL. (1987). Nature 328, 131-136. MEDLINE

Vaccarino FM, Schwartz ML, Raballo R, Nilsen J, Rhee J, Zhou M, Doetschman T, Coffin JD, Wyland JJ and Hung YT. (1999). Nat Neurosci 2, 246-253. Article MEDLINE

Vaziri H and Benchimol S. (1998). Curr Biol 8, 279-282. MEDLINE

van Weering DH, de Rooij J, Marte B, Downward J, Bos JL and Burgering BM. (1998). Mol Cell Biol 18, 1802-1811. MEDLINE

Weng NP, Levine BL, June CH and Hodes RJ. (1996). J Exp Med 183, 2471-2479. MEDLINE

Weng NP, Palmer LD, Levine BL, Lane HC, June CH and Hodes RJ. (1997). Immunol Rev 160, 43-54. MEDLINE

Xu D, Gruber A, Bjorkholm M, Peterson C and Pisa P. (1999). Br J Cancer 80, 1156-1161. Article MEDLINE

Yamaguchi Y, Nozawa K, Savoysky E, Hayakawa N, Nimura Y and Yoshida S. (1998). Exp Cell Res 242, 120-127. MEDLINE

Zhu J, Wang H, Bishop JM and Blackburn EH. (1999). Proc Natl Acad Sci USA 96, 3723-3728. Article MEDLINE

Zhu X, Kumar R, Mandal M, Sharma HW, Dhingra U, Sokoloski JA, Hsiao R and Narayanan R. (1996). Proc Natl Acad Sci USA 93, 6091-6095. MEDLINE

Figure 1 Telomerase activity is strictly regulated during cortical and cerebellar maturation. (a) Telomerase activity was assayed using the TRAPeze-ELISA detection kit (Oncor) on 1 g of proteins extracted from murine cortex, cerebellum and spleen at various stages of development. Standard deviations were calculated from independent experiments performed on three different mice separately. PN1, PN8, PN16: tissues obtained from mice at 1, 8, 16 days respectively after birth. (b) Samples of a typical experiment shows the characteristic processive 6-base pair ladders upon polyacrylamide gel electrophoresis. The lack of detection of telomerase activity in adult cortex and cerebellum is confirmed by the amplification of the 36 bp internal control. 1, 2, 3, 4 and 5: TRAP assay performed on extracts from cortex obtained from E15, PN1, PN8, PN16 and adult mice respectively; 6, 7, 8: TRAP assays performed on cerebellum from PN8, PN16 and adult mice respectively. (c) Decreasing amounts of extracts obtained from PN8 mouse cortex and cerebellum were subjected to TRAP assays. The figure shows the decreasing ELISA values and the corresponding DNA ladders visualized using PAGE

Figure 2 Telomerase activity is down-regulated during neuronal differentiation of cortical cultures. (a) Cortical cultures were obtained from E15 C57BL/6 embryos using a standard protocol and cultured in defined medium. Telomerase activity was assessed at days (D) 1, 4, 7 and 14 after platting using TRAP-ELISA assay on 125 ng of proteins. Standard deviations were calculated from four independent experiments. Astrocytes were purified from cortical cultures of neonatal C57BL/6 mice. (b) Samples of a typical experiment showed the characteristic processive 6-base pair ladder upon polyacrylamide gel electrophoresis. RNase treatment completely abolished the ladders without affecting internal 36 bp control amplification demonstrating thus the specificity of the telomerase detection method. (c) RNA-PCR analysis of mTERT and mGAPDH. Mouse TERT mRNA progressively decreased during neuronal differentiation of cortical cultures

Figure 3 FGF2 induces the proliferation of E15 nestin-positive cortical precursors. (a) [³H]thymidine incorporation after 2 days in vitro of treatments with increasing concentrations of either NGF or FGF2 compared to untreated control cultures. Experiments were performed in sextuplicates. (b) Immunophenotyping of cortical cell cultures after 4 days in vitro using either mAb against MAP2 or nestin and staining using Texas red-coupled secondary antibodies. Nuclei were stained using DAPI. (c) Respective percentages of cells expressing nestin or MAP2 after 4 days in vitro (DIV) in growth factors-treated or -untreated cultures. Increasing of nestin-positive cell percentage in FGF2-treated cultures was associated with a subsequent decrease of percentage of MAP2-positive neurons. The figure also shows the respective percentages of MAP-2 and nestin positive cells in untreated control cultures after 1 day in vitro. Note the decreasing percentage of nestin positive cells from 1 day to 4 days in vitro in untreated control

Figure 4 FGF2 up-regulates telomerase activity in primary cortical cultures. (a) Immediatly after platting, cortical cells were exposed to increasing concentration of FGF2 and NGF in defined medium TRAP-ELISA was performed after 4 days in vitro on 125 ng of cell extracts. Similar experiments were performed using 40 ng/ml of NGF or FGF2 on primary mouse astrocyte cultures. Standard deviations were calculated from two independent experiments. (b) Decreasing amounts of cell extracts obtained from cortical cultures treated or not with 40 ng/ml of FGF2 were subjected to TRAP assay, the figure shows the ELISA values and the DNA ladders visualized using PAGE

Figure 5 FGF2 did not increase mRNA level in cortical cultures compared to that of mGAPH. (a) Typical mRNA PCR analysis showed that mTERT mRNA stayed constant, after 1 day of treatment with an increasing concentration of FGF2 and NGF. (b) Decreasing amounts of cDNA obtained from untreated control and cultures treated by 40 ng/ml of FGF2 or NGF were subjected to PCR amplication. The figure shows the linearity of the assay and the lack of significant increase of mTERT mRNA level in FGF2 treated culture

Figure 6 Zidovudine inhibits both telomerase activity and proliferation of cortical precursors. (a) Immediatly after platting, cortical cells from E15 C57BL/6 embryos were exposed to FGF2 (40 ng/ml) with increasing concentration of AZT (15 and 50 M) in defined medium. After 4 days, cells were harvested and a TRAP ELISA was performed. (b) Samples of a typical experiment show the characteristic 6-base pair ladders upon PAGE. (c) [³H]thymidine incorporation after 2 days in vitro. Experiments were performed in 6-plicates. (d) Immunophenotyping of cortical cells after 4 days in vitro (DIV) exposed to FGF2 with or without AZT using either mAb against MAP2 or nestin. Nuclei were stained using DAPI