Steroid hormone receptors and steroid action in rat glial cells of the central and peripheral nervous system1

PII:

J. Steroid Biochem. Molec. Biol. Vol. 65, No. 1±6, pp. 243±251, 1998 # 1998 Elsevier Science Ltd. All rights reserved Printed in Great Britain S0960-0760(97)00191-X 0960-0760/98 $19.00 + 0.00

Steroid Hormone Receptors and Steroid Action in Rat Glial Cells of the Central and Peripheral Nervous System I. Jung-Testas* and E. E. Baulieu INSERM U 33, University Paris XI, 80 Rue du General Leclerc, 94276 Le Kremlin-BiceÃtre Cedex, France

The nervous system is a target for sex steroid hormones which have profound actions on the growth, maturation, differentiation and functioning of brain cells. We found that some steroids, termed ``neurosteroids'', are synthesized within the brain by glial cells. The term ``neurosteroids'' designates their site of synthesis Ð the nervous system, either de novo from cholesterol or from steroid hormone precursors. The biological effects of steroid hormones are mediated by speci®c high-af®nity intracellular receptors, which, after hormone binding, function as activated transcription factors. The presence of such receptors was shown in primary cultures of oligodendrocytes and astrocytes, derived from forebrains (CNS), and in Schwann cells, derived from sciatic nerves (PNS), of newborn rats. In glial cells of the CNS, progesterone-, glucocorticoid-, estrogen and androgen-receptors (PR, GR, ER, AR) were demonstrated and of these receptors, only PR was estrogen-inducible. In glial cells of the PNS, the presence of PR and ER was shown, but the PR in Schwann cell cultures was not inducible by estrogen treatment. Different effects of steroids on glial cell growth and differentiation during primary culture were observed. In particular, a striking increase of myelin-speci®c proteins such as myelin basic protein (MBP) and cyclic nucleotide phosphodiesterase (CNPase) was observed when oligodendrocytes, the myelinating glial cells of the CNS, were cultured in the presence of progesterone, as determined by indirect immuno¯uorescence staining and immunoblotting. Insulin also increases MBP and CNP-ase in oligodendrocytes and the combined treatment (insulin + progesterone) promotes a strong synergistic stimulation (14-fold increase) of myelin protein expression. Estradiol also increases MBP- and CNPase expression in oligodendrocytes, although to a lesser extent than progesterone. In the search for optimal stimulation of myelin-protein expression, several progesterone analogues were tested and the results are discussed. # 1998 Elsevier Science Ltd. All rights reserved. J. Steroid Biochem. Molec. Biol., Vol. 65, No. 1±6, pp. 243±251, 1998

INTRODUCTION

as, for example, the activation of lordosis re¯exes in female rats by estradiol[1]. Progesterone also plays an important role in the brain and pituitary, where it regulates female reproductive behavior and gonadotropin secretion [2]. Previous observations from our group have shown that dehydroepiandrosterone and pregnenolone, the precursor of progesterone accumulate in rat brain independently of adrenal and gonadal sources, as shown by the persistence of these steroids in the brain after gland ablation or pharmacological suppression [3]. This observation led us to search whether these steroids, called ``neurosteroids'', might be synthesized in the brain, either de novo from cholesterol or by in situ metabolism of steroid hormone precursors [4, 5]. Steroidogenesis begins with the con-

The nervous system is a target for sex steroid hormones which have profound actions on the growth, maturation, differentiation and functioning of brain cells. In the central nervous system (CNS), progestins and estrogens act in concert to promote changes in sexual behavioral responses and physiology. The cellular mechanisms of the action of steroid hormones have been studied most extensively in rodents such Proceedings of the 13th International Symposium of the Journal of Steroid Biochemistry & Molecular Biology ``Recent Advances in Steroid Biochemistry & Molecular Biology'' Monaco, 25±28 May 1997. *Correspondence to I. Jung-Testas. Tel: 331 495 91803; Fax: 331452 11940. 243

244

I. Jung-Testas and E. E. Baulieu

version of cholesterol to pregnenolone by the sidechain cleavage cytochrome P450scc. This enzymatic step is characteristic of the steroidogenic cells of endocrine glands including ovaries, testes and adrenal glands. With the use of primary cultures of glial cells, established from newborn rat forebrains and composed of oligodendrocytes and astrocytes, we found that the cytoplasm of both cell types was speci®cally immunostained with antibodies to bovine adrenal cytochrome P450scc [6]. After incubation of these cells with labeled mevalonate as steroid precursor, or labeled cholesterol linoleate, incorporated into reconstituted low-density lipoproteins, the biosynthesis of pregnenolone and progesterone was demonstrated de®nitely indicating that brain cells can perform steroid biosynthesis [6±8]. Considering that progesterone is synthesized within the brain cells, it was of interest to investigate the presence of progesterone receptors in glial cells. Indeed, biological effects of steroid hormones are mediated by speci®c receptor proteins which, after hormone binding, function as activated transcription factors that bind to speci®c nuclear response elements to modulate transcription of target genes [9, 10]. The presence of such steroid hormone receptors within different regions of the brain has been described by several groups [2, 11], and we have demonstrated 4 classes of steroid receptors in forebrain cell cultures, progesterone, estrogen, androgen and glucocorticoid receptors (PR, ER, AR, GR). Estrogen treatment of the cultures signi®cantly increased PR, whereas ER, AR and GR remained unchanged. Moreover, the PR-induction was most signi®cant in glial cells established from female rat pups [12, 13]. In agreement with the presence of intracellular receptors, brain glial cells are sensitive to the action of steroid hormones, as seen by changes in cell growth, morphology and differentiation. For instance, progesterone increases the synthesis of myelin proteins in oligodendrocytes during early differentiation [13]. In the CNS, myelin is formed by oligodendrocytes. During myelinogenesis, oligodendrocytes form numerous cytoplasmic extensions, each of which enwraps axonal segments, thereby permitting fast conduction of nerve impulses along axons [14]. In rodents, central myelin is composed of about 70% lipids, among which galactocerebroside (GalC) is a speci®c marker for oligodendrocytes [15, 16], and of about 30% proteins. The myelin proteins are mainly composed of proteolipid protein, myelin basic protein (MBP), 2',3'-cyclic nucleotide3'-phosphodiesterase (CNPase) and myelin associated glycoproteins [15, 17]. In the peripheral nervous system (PNS), the myelinating glial cells are Schwann cells. During development, Schwann cells proliferate and migrate along axons, surrounding them to elaborate the myelin sheaths [18]. If the nerve is injured, Schwann cells again proliferate to restore the integrity of the myelin

sheaths and aid regeneration of the lesioned axons [19, 20]. As we had shown that estradiol increases and progesterone inhibits glial cell proliferation in the CNS [13], it was possible that steroid hormones could also in¯uence the myelination process in the PNS, as for example by increasing Schwann cell proliferation. This was tested in primary cultures of Schwann cells, prepared from neonatal rat sciatic nerves. Schwann cells can be easily grown and puri®ed in culture, and pure secondary cultures preserve basic Schwann cell functions [21, 22]. Recent studies with such Schwann cells have shown that estradiol, in the presence of forskolin or dbc-AMP, stimulates cell proliferation. The presence of intracellular receptors for estrogen and for progestin was also shown in Schwann cell cultures [23, 24]. However, in contrast to oligodendrocytes, which differentiate and produce myelin proteins in cell culture in the absence of neurons [25], Schwann cells in culture remain undifferentiated. Therefore glial cells from the CNS are a promising tool to study a wide range of cell biological questions, especially the interaction of steroids and growth factors on proliferation, differentiation and on myelin protein synthesis, and for the investigation of pharmacological drugs which could promote myelinating processes.

STEROID HORMONE RECEPTORS IN CULTURED GLIAL CELLS

Glial cells of the CNS Primary cultures established from newborn rat cerebral hemispheres contain two main cell types: oligodentrocytes and astrocytes. Both cell types can be identi®ed with speci®c antibodies against galactocerebroside (GalC), myelin basic protein (MBP) and 2',3'-cyclic nucleotide-3'phosphodiesterase (CNPase) for oligodendrocytes [16, 17, 25], and against glial ®brillary acidic protein (GFAP) for astrocytes [26]. Primary glial cell cultures can be maintained in vitro for several months, however, glial cells proliferate actively only during the ®rst weeks [7]. Ligand binding experiments revealed the presence of PR, GR, ER and AR in cytosol- and nuclear fractions, after labeling cells at 378C with the appropriate ligand (whole cell assay). PR and GR were further analyzed by gradient ultracentrifugations, both receptors displayed a 9 S form in the cytosol and a 4±6 S form in nuclear high-salt extracts [12, 13]. In the rat brain and pituitary, estradiol has inductive effects on the concentration of PR in in vivo experiments [27], and a similar estrogen-inducibility of PR was observed in glial cell cultures. The induction was speci®c for PR, since levels of GR, ER, and AR in parallel experiments were unaffected by the estrogen-treatment [13]. Moreover, the antiestrogen

Steroid hormone receptors and steroid action

Fig. 1. Whole cell assay of progesterone receptor in brain glial cells. Effect of estradiol and anti-estrogen effect of tamoxifen. Primary cultures of new born rat glial cells were treated (or not) from days 10±27 with 50 nM E2, with 50 nM E2 plus 100 nM tamoxifen (Tam), or with 100 nM Tam alone. Control cells (no H) received vehicle only. Cells were then incubated with 5.0 nM [3 H]-Org 2058 for 1 h at 378C with or without 1 mM radioinert progesterone. After washing the cells thoroughly with ice-cold PBS, they were homogenized in PGW-buffer (10 mM potassium phosphate, 10% glycerol (v/ v), 20 mM tungstic acid, PH 7.8), centrifuged at 1000 for 20 min, and the nuclear pellet extracted in PGW containing 0.4 M KCl. Speci®c binding, (difference between total- and nonspeci®c binding, determined in the presence of competitor), was expressed as bound [3 H]-Org 2058 per mg nuclear protein. Each column represents the mean values of three separate experiments. The SEMs were less than 10%.

tamoxifen completely abolished the estrogen-induction of PR (Fig. 1). Sex differences on PR and ER levels in certain regions of the brain have been reported [28], as well as sex differences in PR-induction after estrogentreatment of male and female rats [29]. In selecting cultures from newborn male or female pups, the estrogen-inducibility of PR was substantially increased in female cultures (Fig. 2). Immunostaining of PR with monoclonal anti-PR antibodies revealed the presence of progestin receptors in both oligodendrocytes and astrocytes. In estrogen-deprived cultures, only weak PR staining can be observed, however; after estrogen treatment of the cells, PR-staining increased signi®cantly and was more intense in oligodendrocytes than in astrocytes [12]. Glial cells of the PNS The myelinating glial cells of the PNS are Schwann cells. They surround all peripheral nerve axons and

245

Fig. 2. Measure of progesterone receptor in glial cell cultures from male or female offsprings. Cultures, established from male or female pups at day 1 of birth, were treated with 50 nM E2 (or not), and speci®c PR-binding was measured after incubating cells with [3 H]-Org 2058, 2.0 or 5.0 nM, in absence or presence of competitor, as described in Fig. 1. q E2 treatment. Reproduced with percontrol cultures mission from Jung-Testas et al.[13].

elaborate the myelin sheath of myelinated nerve ®bers. Schwann cells proliferate during development and also after nerve injury to restore the integrity of the myelin sheath. Little was known about the involvement of steroid hormones in Schwann cell proliferation. Studies on Schwann cell tumors indicated the presence of estradiol binding sites in neuro®bromas, which usually develop at the time of puberty and enlarge with pregnancy [30±32]. The antiestrogen tamoxifen has been shown to reduce the incidence of peripheral nerve tumors in an animal model [30]. Thus the existence of estrogen receptors in Schwann cells was possible, and primary cultures from sciatic nerves of newborn rats were established for these studies. Schwann cells in culture have a characteristic spindle-shaped morphology with an oval nucleus and long processes. They can be characterized by immuno¯uorescence staining with antibodies against the neuroectodermal S-100 antigen, a speci®c marker for

246

I. Jung-Testas and E. E. Baulieu

Schwann cells, which was found at high levels in peripheral nerve [33]. After puri®cation, Schwann cells can be multiplied easily, they proliferate rapidly in the presence of elevated levels of cAMP and speci®c peptide growth factors [34]. The presence of speci®c estrogen binding in such cultures was shown by whole cell uptake of radiolabeled estradiol. The binding was saturable and could be increased by elevating intracellular levels of cAMP. The presence of ER in Schwann cells was con®rmed by immuno¯uorescence staining [23]. Most of the ER-staining was seen in the cytoplasm, but exposure of the cells to estradiol (200 nM) resulted in a strong increase of nuclear staining, suggesting that the receptor is nuclear as observed for the estrogen receptor in classical target cells. In addition to ER, the presence of progestin receptors was demonstrated in puri®ed Schwann cell cultures by binding of the selective ligand [3 H]-Org 2058 and by immuno¯uorescence staining with speci®c PR-antibodies [24]. Interestingly, as shown in Fig. 3, the progesterone receptor in Schwann cell cultures is not inducible by estrogen treatment of the cells, as observed in glial cells of the CNS. Similar to the glial cells of the CNS, Schwann cells can also synthesize neurosteroids, including pregnenolone progesterone and their reduced metabolites [35]. However, in pure Schwann cell cultures, progesterone synthesis could not be observed. In contrast, Schwann cells isolated from embryonic

Fig. 3. Binding analysis of [3 H]-Org 2058 to progesterone receptor in cultured rat Schwann cells. Schwann cells were incubated for 1 h at 378C with increasing concentrations of [3 H]-Org 2058 in presence or absence of 1 mM unlabeled progesterone. Speci®c PR-binding sites were measured as described in Fig. 1. Results are expressed as dpm bound per 1  106 cells. ± * ± Schwann cells cultured for 2 weeks prior the PR measurement in the presence of 100 nM estradiol. ± w ± Control cultures. Experiments were done in triplicate (3 measures for each experiment), mean values2S.D. are indicated. Reproduced with permission from Jung-Testas et al. [24].

rat DRG-explants, which have been co cultured with sensory neurons, can synthesize progesterone [36], suggesting that neurons induce the biosynthetic pathway of progesterone in Schwann cells. It is thus possible that neuronal signals are necessary for the expression of Schwann cell-speci®c factors regulating the estrogen-responsiveness of the progestin receptor in Schwann cell cultures.

GLIAL CELLS ARE TARGETS FOR STEROID HORMONES

Glial cell proliferation In agreement with the presence of intracellular receptors for steroids, glial cells are sensitive to the actions of steroid hormones. For instance, estradiol stimulates growth of glial cells from the CNS, oligodendrocytes and astrocytes, whereas progesterone inhibits cell proliferation. Interestingly, simultaneous treatment of the cells with progesterone (100 nM) and the antagonist RU486 (200 nM) resulted in a complete abolition of the growth inhibition [13]. Schwann cell proliferation is also substantially increased by estradiol, but only in the presence of forskolin, a reversible activator of adenylate cyclase, or dbc-AMP. The mitogenic effect of estradiol is already maximal at low nanomolar concentrations and can be blocked by the anti-estrogen ICI-164,384. In contrast to glial cells of the CNS, progesterone did not in¯uence Schwann cell growth [23]. The synergism between estradiol and cAMP might correspond to a general mechanism by which estradiol and cAMP activate cell division in concert, or estradiol could modulate the sensitivity of Schwann cells to cAMPdependent growth factors [34]. We have identi®ed the insulin-like growth factor IGF-1 as a potent mitogen for Schwann cells in the presence of dbc-AMP or forskolin. Indeed, Schwann cells express the receptor for IGF-1, whose number increases if the cells are cultured for several days in the presence of forskolin[37]. As the proliferation of Schwann cells plays an important role during the regeneration of peripheral nerves after injury, the mitogenic actions of estradiol and IGF-1 may have bene®cial effects on myelin repair. Glucocorticoids are also mitogenic for Schwann cells and potentiate the actions of growth factors [38], and it is likely that steroid hormones exert their trophic effects by acting in concert or synergy with peptide growth factors such as IGF-1, FGF, PDGF or cytokines. Steroids induce changes in brain cell morphology and adhesiveness Brain glial cells in primary culture undergo dramatic morphologic changes after several days of culture in the presence of progesterone (200 nM). Both oligodendrocytes and astrocytes develop long, straight

Steroid hormone receptors and steroid action

247

Table 1. Effects of hormones and antihormone on glial cells adhesiveness Compounds added Control Progesterone (100 nM) RU486 (100 nM) Progesterone (100 nM) + RU486 (200 nM) D5 Pregnenolone (100 nM) Pregnanediol (100 nM) Estradiol (100 nM) Dexamethasone (100 nM)

Attached cells (Index*) 100 69 111 70 96 104 130 99

Primary glial cells of the CNS were subcultured at day 8 and replated in Dulbeccos Modi®ed Eagle medium containing 5% steroid-free calf serum, in the absence or presence of the indicated compounds. Plated cells = 3.1  105/Petri dish (60 mm). *100 = 2.1  105 cells counted 30 h later.

membrane processes, much more abundant than in control cultures. To our surprise, the progesterone antagonist RU486 could not prevent these morphologic changes of cell shape which could indicate a non genomic effect of progesterone, most likely by binding to membrane receptors. Indeed, the possible existence of membrane receptors for estrogen, progesterone and testosterone has been described in the rat brain [39]. Estradiol, at the same nanomolar concentrations as progesterone, induced similar morphological changes in both cell types, but the effect was less dramatic. The adhesiveness of cells to the surface of culture dishes is also in¯uenced by steroid hormones, again suggesting direct actions on the cell membrane. For example, estradiol increases and progesterone decreases plating ef®ciency as shown in Table 1. Again, the progesterone antagonist RU486 could not prevent the negative effect of progesterone, and had no effect by itself. The neurosteroids pregnenolone and pregnanediol had no in¯uence on cell adhesiveness. Comparable observations were made in Schwann cell cultures (A. Do Thi and I. Jung-Testas, unpublished work, 1997), suggesting steroid actions at the plasma membrane level in both types of glial cells. Examples of non-genomic mechanisms of action of steroid hormones are known and have been extensively reviewed by Baulieu et al. [40]. PROGESTERONE INCREASES EXPRESSION OF MYELIN PROTEINS IN OLIGODENDROCYTES

As mentioned above, among myelin-speci®c markers, MBP, CNPase and GalC are expressed early in oligodendrocytes. During the ®rst days of primary culture (day 1 corresponding to the day of birth), the expression of these markers can be followed by speci®c immuno¯uorescence staining. GalC is

Fig. 4. Expression of MBP, CNPase and GalC in rat oligodendrocytes in primary culture. Glial cells were plated at day 1 (day of birth) on PLL-treated glass coverslips and cultured in DME-medium containing 10% charcoal-treated calf serum, in the absence, or presence of insulin, 4.5 mg/ml, or insulin 4.5 mg/ml + progesterone (500 nM), or insulin 4.5 mg/ ml + estradiol (500 nM). Hormones were added daily, media were changed every 2 or 3 d. At day 9, immuno¯uorescence staining of MBP, CNPase and GalC was done and the number of antigen-expressing cells per 2000 oligodendrocyte-like cells are indicated. The median of 3 independent measures are shown. Modi®ed from Jung-Testas et al.[41].

expressed ®rst in oligodendrocytes (day 2), followed by MBP (days 4±6) and CNPase (days 6±8). The treatment of glial cells from day 1 onwards with progesterone increased MBP- and CNPase expression in oligodendrocytes. Estradiol was less effective, but neither hormone had an in¯uence on GalC expression [13, 41, 42]. Insulin is known to increase MBP-expression in oligodendrocytes [43] and the combined treatment of cells with insulin and progesterone resulted in a strong synergistic stimulation of MBP- and CNPase expression. Here again, estradiol had similar, but less potent effects on the synthesis of the myelin proteins (Fig. 4). The stimulation of protein synthesis by progesterone was also shown by immunoblotting analysis of CNPase. Single bands of the protein could be detected from day 11 of primary culture onwards, and its synthesis gradually increased thereafter. In the

248

I. Jung-Testas and E. E. Baulieu

Fig. 5. Western immunoblot of CNPase protein from rat brain glial cells after different days of primary culture. Samples containing 200 mg of glial cell protein were resolved by 10% SDS-PAGE and anti-CNPase antibodies. Puri®ed rat myelin (M) was analyzed in parallel as a control. Primary cultures were prepared at day 1 (day of birth) and the cells treated (+P) or not (C) with progesterone (200 nM) from day 1 onwards until day 16. Reproduced with permission from Jung-Testas et al. [42].

presence of progesterone, CNPase expression is accelerated and the intensity of the bands had clearly increased (Fig. 5). These trophic effects of progesterone on myelin synthesis may have clinical implications, and it is certainly worth considering the use of steroid compounds or drugs to treat lesions and diseases of myelin. Our brain cell culture system provides a powerful research tool to explore the relative contribution of such compounds in myelin synthesis during development. Similar studies cannot be done in Schwann cell cultures, since Schwann cells remain undifferentiated and do not express myelin proteins under usual culture conditions. However, recent in vivo studies on peripheral nerve repair in male mice have shown that progesterone also plays an important role in myelinization of sciatic nerves after freeze lesions [36].

Effect of progesterone analogues on myelin-protein synthesis Three orally active progestogens whose pharmacological pro®le and structure resembles the natural hormone progesterone were tested: megestrol acetate (17-hydroxy-6-methylpregna-4,6-diene-3,20-dione acetate), chlormadinone acetate (17-acetyloxy-6chloro-pregna-4,6-diene-3,20-dione acetate) and drospirenone (6b,7b,15b,16b,-dimethylene-3-oxo17a-pregn-4-ene-21,17-carbolactone). They were kindly provided by Dr. E. Schillinger, Schering, Berlin. The relative binding af®nity of the three progestogens to cytosolic and nuclear PR of rat brain glial cells was estimated, in comparison to progesterone and estradiol, the latter being a ``negative'' control. The results are summarized in Table 2 and Fig. 6. At a low competitor concentration (500 nM),

Table 2. Speci®c binding of [3 H]-Org 2058 to PR in cytosol and nuclear extract of rat glial cells in absence or presence of competitor Cytosol Competitor None Progesterone Megestrol±Ac. Chlormadinone±Ac. Drospirenone Estradiol

Nuclear extract

500 nM

2 mM

500 nM

2 mM

100 71 82 85 73 101

100 49 69 79 64 99

100 57 81 86 69 100

100 43 58 67 62 101

Glial cells, 3 weeks after primary culture, were incubated for 1 h at 378C with 3 H-Organon 2058, 6 nM, in absence (none) or presence of competitor (500 nM or 2 mM). Speci®c PR-binding was determined in cytosol and nuclear extract by subtracting counts obtained in the presence of competitor. Data are presented as percent speci®c binding in the absence of competitor. Experiments were done in triplicate (3 different primary cultures for each experiment), mean values are indicated. The SEMs were less than 10%.

Steroid hormone receptors and steroid action

249

Table 3. Effect of progesterone, megestrol acetate and RU486 on MBP- and CNP-ase expression in rat oligodendrocytes Day 12 of primary culture Compounds added Control Prog 200 nM Prog 1 mM Meg 200 nM Meg 1 mM RU486 1 mM Prog 200 nM + RU486 1 mM Meg 200 nM + RU486 1 mM

MBP+ 126 395 420 380 510 180 384 373

CNPase+ 83 280 308 268 405 153 250 220

Glial cells were prepared at postnatal day 1, plated on PLL-treated glass coverslips and cultured in DME-medium containing 10% charcoal-treated calf serum and insulin (4.8 mg/ml), in the absence (control), or presence of progesterone (Prog), (alone or together with RU486) megestrol acetate (Meg), (alone or together with RU486), and RU486, at the indicated concentrations. At day 12, immuno¯uorescence staining of MBP and CNP-ase was performed. Labeled and unlabeled oligodendrocytes were counted and the number of antigen-expressing cells per 2000 oligodendrocytes are indicated. Data correspond to 3 different experiments, the mean values for the 3 determinations are shown. The SEMs were less than 10%. Fig. 6. Speci®c binding of [3 H]-Org 2058 to the progesterone receptor in nuclear extract of rat brain glial cells, in the absence (none) or presence of progesterone (Prog), megestrol acetate (Meg±Ac), or estradiol (E2) as competitor, at 0.5 or 2.0 mM concentration. PR-binding was determined by whole cell assay as described in Fig. 1, the results are presented as percent speci®c binding in the absence of competitor. Experiments were done in triplicate (3 different primary cultures for each experiment). The SEMs were less than 10%. 3

only progesterone displaced [ H]-Org 2058-binding signi®cantly in the nuclear fraction, but at a higher competitor concentration, the 3 progestogens are almost as ef®cient as progesterone in displacing the ligand binding. As expected, estradiol does not compete for PR. Yet, the similar binding af®nities of the three progestogens to the glial PR are not accompanied by a corresponding agonistic action on myelin protein synthesis. Chlormadinone acetate and drospirenone have little effect (data not shown), but megestrol acetate strongly induced MBP- and CNPase expression and the induction is dose-dependent (Table 3). The antagonist RU486 does not inhibit the induction of myelin protein synthesis observed in the presence of progesterone and megestrol acetate, and it has a slight agonistic action by itself. However, RU486 has high af®nity for the brain glial PR [13]. Taken together, these results indicate that high binding af®nity of compounds to the glial PR does not necessarily correspond to agonistic action. Especially interesting is the trophic effect of megestrol acetate on myelin-protein synthesis during early development of oligodendrocytes (Fig. 7), and further investigations should assess the physiological signi®-

cance of our ®nding, which might have clinical implications in the treatment of diseases and lesions of myelin.

CONCLUSIONS

The preceding sections have reviewed the results obtained from our studies on rat glial cells in primary culture. This cell culture system unequivocally demonstrates the biosynthesis of neurosteroids in myelinating glial cells of the CNS and the PNS. Considering that progesterone can be synthesized by glial cells, this sex steroid hormone, which is produced and secreted by the ovaries and adrenal glands, is also a neurosteroid. Estradiol a priori cannot be synthesized de novo from cholesterol or pregnenolone within the brain since the cytochrome P45017a, the obligatory step in formation of precursors of androgens and estrogens, seems to be missing in the nervous system [44]. However, both progesterone and estradiol have multiple effects on glial cells and the demonstration of the corresponding intracellular hormone receptors suggests, at least in part, a classic intracellular mechanism of action in an autocrine/paracrine manner. Especially interesting is the trophic effect of progesterone and the progesterone analogue megestrol acetate on central myelin protein synthesis, providing compelling support for an important role for progestins in the myelinating process. Thus our brain cell culture system provides an essential research tool for studying a wide range of cell biological questions, and to explore the relative contributions of steroid hor-

250

I. Jung-Testas and E. E. Baulieu

Fig. 7. Induction of MBP in rat oligodendrocytes by megestrol acetate. Glial cells were plated at day 1 on PLL-treated glass coverslips and cultured in DME-medium containing 10% charcoal-treated calf serum and insulin (4.5 mg/ml). Megestrol acetate (200 nM) was added daily (right), control cultures received vehicle only (left). At day 8, cells were ®xed and immunostained for MBP. Representative microscope ®elds are represented. Most ®elds of control cultures (left) contained no MBP-expressing oligodendrocytes (400).

mones and growth factors in myelin synthesis during development. AcknowledgementsÐWe thank C. Legris for ef®cient and excellent assistance in preparing the manuscript, J.C. Lambert and L. Outin for the preparation of ®gures and photographs and B. Delespierre for skilful technical assistance. This work was supported by INSERM.

REFERENCES 1. Green R., Luttge W. G. and Whalen R. E., Induction of receptivity in ovariectomized female rats by a single intravenous injection of estradiol-17b. Physiology 5 (1970) 137±142. 2. McEwen B. C., Biegon A., David P. G., Krey L. C., Luine V. N., McGinnis Y., Paden C. M., Parsons B. and Rainbow T. C., Steroid hormone: Humoral signals which alter brain cell properties and functions. Recent Prog. Horm. Res. 38 (1982) 41±83. 3. Baulieu, E. E., Robel, P., Vatier, O., Haug, M., Le Goascogne, C., Bourreau, E. Neurosteroids: Pregnenolone and dehydroepiandrosterone in the brain. In Receptor±Receptor Interactions, Vol. 48. MacMillan Press Ltd., Basingstoke, 1987, pp. 89±104. 4. Robel P. and Baulieu E. E., Neuro-steroids: 3beta-hydroxydeltasub 5-derivatives in the rodent brain. Neurochem. Int. 7 (1985) 953±958. 5. Baulieu E. E., Neurosteroids: A new function in the brain. Biol. Cell 71 (1991) 3±10.

6. Jung-Testas I., Alliot F., Pessac B., Robel P. and Baulieu E. E., Localisation immunocytochimique du cytochrome P450scc dans les oligodendrocytes de rat en culture. C. R. Acad. Sci. Paris 308 (1989) 165±170. 7. Jung-Testas I., Hu Z. Y., Baulieu E. E. and Robel P., Neurosteroids: Biosynthesis of pregnenolone and progesterone in primary cultures of rat glial cells. Endocrinology 125 (1989) 2083±2091. 8. Jung-Testas I., Weintraub H., Dupuis D., Eychenne B., Baulieu E. E. and Robel P., Low density lipoprotein-receptors in primary cultures of rat glial cells. J. Steroid Biochem. Mol. Biol. 42 (1992) 597±605. 9. Evans R. M., The steroid and thyroid hormone receptor superfamily. Science 240 (1988) 889±895. 10. O'Malley B. W. and Tsai M. J., Molecular pathways of steroid receptor action. Biol. Reprod. 46 (1992) 163±167. 11. Kato, J., Progesterone receptors in brain and hypophysis. In Current Topics in neuroendocrinology, Vol. 5. Springer-Verlag, Berlin, 1985, pp. 31±81. 12. Jung-Testas I., Renoir J. M., Gasc J. M. and Baulieu E. E., Estrogen-inducible progesterone receptor in primary cultures of rat glial cells. Exp. Cell Res. 193 (1991) 12±19. 13. Jung-Testas I., Renoir M., Bugnard H., Greene G. L. and Baulieu E. E., Demonstration of steroid hormone receptors and steroid action in primary cultures of rat glial cells. J. Steroid Biochem. Mol. Biol. 41 (1992) 621±631. 14. Ritchie, J. M., Physiological basis of conduction in myelinated nerve ®bers. In Myelin, ed. P. Morrell. Plenum Press, New York, 1984, pp. 117±145.

Steroid hormone receptors and steroid action 15. Norton, W. T., Biochemistry of myelin. In Advances in Neurobiology, ed. S. G. Waxman and J. M. Ritchie. Raven Press, New York, 1981, pp. 93±121. 16. Raff M. C., Mirsky R., Fields K. L., Lisak R. P., Dorfman S. H., Silberberg D. H., Gregson N. A., Leibowitz S. and Kennedy M. C., Galactocerebroside is a speci®c cell-surface antigenic marker for oligodendrocytes in culture. Nature 274 (1987) 813±816. 17. Lees M. B., Brostof, S. W., Proteins of myelin. In Myelin, 2nd Edn., ed. P. Morrell. Plenum Press, New York, 1984, pp. 197±217. 18. Asbury A. K., Schwann cell proliferation in developing mouse sciatic nerve. J. Cell Biol. 34 (1967) 735±743. 19. Salzer J. L. and Bunge R. P., Studies of Schwann cell proliferation. I. An analysis in tissue culture of proliferation during development, Wallerian degeneration, and direct injury. J. Cell Biol. 84 (1980) 739±752. 20. Fawcett J. W. and Keynes R. J., Peripheral nerve regeneration. Annu. Rev. Neurosci. 13 (1990) 43±60. 21. Brockes J. P., Fields K. L. and Raff M. C., Studies on cultured rat Schwann cells. I. Establishment of puri®ed populations from cultures of peripheral nerve. Brain Res. 165 (1979) 105± 118. 22. Porter S., Clark M. B., Glaser L. and Bunge R. P., Schwann cells stimulated to proliferate in the absence of neurons retain full functional capability. J. Neurosci. 6 (1986) 3070±3078. 23. Jung-Testas I., Schumacher M., Bugnard H. and Baulieu E. E., Stimulation of rat Schwann cell proliferation by estradiol: Synergism between the estrogen and cAMP. Dev. Brain Res. 72 (1993) 282±290. 24. Jung-Testas I., Schumacher M., Robel P. and Baulieu E. E., Demonstration of progesterone receptors in rat Schwann cells. J. Steroid Biochem. Mol. Biol. 58 (1996) 77±82. 25. Dubois-Dalcq M., Behar T., Hudson L. and Lazzarini R. A., Emergence of three myelin proteins in oligodendrocytes cultured without neurons. J. Cell Biol. 102 (1986) 384±392. 26. Bignami A., Eng L. F., Dahl D. and Uyeda C. T., Localization of the glial ®brillary acidic protein in astrocytes by immuno¯uorescence. Brain Res. 43 (1972) 429±435. 27. Maclusky N. J. and McEwen B. S., Progestin reseptors in rat brain: Distribution and properties of cytoplasmic progestin binding sites. Endocrinology 106 (1980) 192±202. 28. Rainbow T. C., Parsons B. and McEwen B. S., Sex differences in rat brain oestrogen and progestin receptors. Nature 300 (1982) 648±649. 29. Coirini H. and McEwen B. S., Progestin receptor induction and sexual behavior by estradiol treatment in male and female rats. J. Neuroendocrinol. 2 (1990) 467±472. 30. Jay J. R., MacLaughlin D. T., Badger T. M., Miller D. C. and Martuza R. L., Hormonal modulation of Schwann cell tumors.

31. 32. 33. 34. 35. 36.

37.

38. 39.

40. 41.

42.

43.

44.

251

I. The effects of estradiol and tamoxifen on methylnitrosoureainduced rat Schwann cell tumors. Ann. N. Y. Acad. Sci. 486 (1986) 371±382. Martuza R. L., MacLaughlin D. T. and Ojemann R. G., Speci®c estradiol binding in schwannomas, meningiomas, and neuro®bromas. Neurosurgery 9 (1981) 665±671. Swapp G. H. and Main R. A., Neuro®bromatosis and pregnancy. Br. J. Dermatol. 80 (1973) 431±435. Moore B. W., Perez V. J. and Gehring M., Assay and regional distribution of a soluble protein characteristic of the nervous system. J. Neurochem. 15 (1968) 265±272. Eccleston P. A., Regulation of Schwann cell proliferation: Mechanisms involved in peripheral nerve development. Exp. Cell Res. 199 (1992) 1±9. Akwa Y., Schumacher M., Jung-Testas I. and Baulieu E. E., Neurosteroids in rat sciatic nerves and Schwann cells. C. R. Acad. Sci. Paris 316 (1993) 410±414. Koenig H. L., Schumacher M., Ferzaz B., Do Thi A. N., Ressouches A., Guennoun R., Jung-Testas I., Robel P., Akwa Y. and Baulieu E. E., Progesterone synthesis and myelin formation by Schwann cells. Science 268 (1995) 1500±1503. Schumacher M., Jung-Testas I., Robel P. and Baulieu E. E., Insulin-like growth factor I: A mitogen for rat Schwann cells in the presence of elevated levels of cyclic AMP. Glia 8 (1993) 232±240. Neuberger T. J., Kalimi O., Regelson W., Kalimi M. and De Vries G. H., Glucocorticoids enhance the potency of Schwann cell mitogens. J. Neurosci. Res. 38 (1994) 300±313. Ramirez V. D., Zheng J. and Siddique K. M., Membrane receptors for estrogen, progesterone, and testosterone in the rat brain: Fantasy or reality. Cell. Mol. Neurobiol. 16 (1996) 175±198. Baulieu, E. E., Robel, P., Non-genomic mechanisms of action of steroid hormones. In Non-Reproductive Actions of Sex Steroids. John Willey and Sons, Chichester, 1995, pp. 24±42. Jung-Testas I., Schumacher M., Robel P. and Baulieu E. E., Actions of steroid hormones and growth factors on glial cells of the central and peripheral nervous system. J. Steroid Biochem. Mol. Biol. 48 (1994) 145±154. Jung-Testas I., Schumacher M., Robel P. and Baulieu E. E., The neurosteroid progesterone increases the expression of myelin proteins (MBP and CNPase) in rat oligodendrocytes in primary culture. Cell. Mol. Neurobiol. 16 (1996) 439±443. Saneto R. P., Low K. G., Melner M. H. and de Vellis J., Insulin/insulin-like growth factor I and other epigenetic modulators of myelin basic protein expression in isolated oligodendrocyte progenitor cells. J. Neurosci. Res. 21 (1988) 210±219. Robel P. and Baulieu E. E., Neurosteroids: Biosynthesis and function. Trends Endocrinol. Metab. 5 (1994) 1±8.