- Email: [email protected]
Testosterone and estrogen affect neuronal differentiation but not proliferation in early embryonic cortex of the rat: the possible roles of androgen and estrogen receptors
Neuroscience Letters 281 (2000) 57±60 www.elsevier.com/locate/neulet
Testosterone and estrogen affect neuronal differentiation but not proliferation in early embryonic cortex of the rat: the possible roles of androgen and estrogen receptors Lei Zhang a,*, Yoong H. Chang b, Jeffery L. Barker b, Qian Hu b, Li Zhang c, Dragan Maric b, Bing-Sheng Li d, David R. Rubinow a a
Building 36, 2C02 Behavioral Endocrinology Branch, NIMH, NIH, 9000 Rockville Pike, Bethesda, MD 20892, USA b Laboratory of Neurophysiology, NINDS, NIH, Bethesda, MD 20892, USA c Synaptic Function Unit, NINDS, NIH, Bethesda, MD 20892, USA d Laboratory of Neurochemistry, NINDS, NIH, Bethesda, MD 20892, USA Received 3 December 1999; received in revised form 10 January 2000; accepted 11 January 2000
Abstract We examined the effect of testosterone (T) and 17 beta-estradiol (E) on differentiation and proliferation of cultured neurons from the cortex of 14-day-rat embryos (E14) using immunocytochemistry. We found that the cultures receiving E had signi®cantly more neurons with longer neurites than the control cultures, while both fewer and less differentiated neurons were seen after 24 h of incubation with T. However, neither T nor E changed the number of cells positive for BrdU, a proliferation marker. We also found that the androgen receptor (AR) was markedly expressed in the neurons, whereas the expression of estrogen receptor (ERa ) was barely detectable. These results suggest that E and T differ in effect on differentiation, while neither affect proliferation in early developmental cortex. Furthermore, since the AR is expressed in the cortical neurons by E14, the inhibitory effect of T on differentiation may be receptor-mediated, while the stimulatory effects of estrogen in the cortex do not appear to involve nuclear ERa at this developmental stage. Published by Elsevier Science Ireland Ltd. Keywords: Testosterone; Estrogen; Sexual dimorphism; Cortex; Differentiation; Proliferation
Sex steroids, including testosterone (T) and estradiol (E), help direct the sexual differentiation and cellular proliferation of the brain [6,14]. Estrogen stimulates neural differentiation and modulates neural survival both in vitro and in vivo [4,6]. Similarly, androgens such as T are known to affect cell growth, differentiation, morphology, and survival [5,8]. Further, at least some of the effects of T on neuronal morphology and differentiation are mediated by its conversion to E [10]. The mechanisms by which T and E regulate neuronal differentiation are essentially unknown. If the sexually dimorphic nature of the brain is determined primarily by sex-related differences in sex hormone levels (although see Breedlove [3] and McEwen [14] for contribution of behavioral and genetic factors), one should be able to
* Corresponding author. Tel.: 11-301-402-1398; fax: 11-301402-1565. E-mail address: [email protected] (L. Zhang)
demonstrate a differential effect of T and E on cellular differentiation. To test this hypothesis, we ®rst developed an embryonic culture system in which we could observe both neuronal proliferation and differentiation. Second, we examined the effects of T and E on proliferation and differentiation, measuring cell number, bromodeoxyuridine (BrdU) labeling (a proliferation marker), and neurite length in cells labeled with the class III neuron-speci®c b -tubulin protein (TUJ1). Finally, we examined the expression and distribution of androgen receptors (AR) and estrogen receptors (ER) in the cultures by immunocytochemistry to determine the possible mechanism for any observed responses to the sex steroids. Cultured neurons were derived from the cerebral cortex of Sprague±Dawley rat embryos (Taconic) on embryonic day 14 (E14), the time when the ®rst postmitotic neurons of the cortex are beginning to differentiate, cell proliferation is predominant, and sex-related differences in the levels of T in the circulation have not appeared [9]. The cerebral hemisphere was dissected and gently triturated with
0304-3940/00/$ - see front matter. Published by Elsevier Science Ireland Ltd. PII: S03 04 - 394 0( 9 9) 00 94 2- 8
58
L. Zhang et al. / Neuroscience Letters 281 (2000) 57±60
a pipette, so that most of the tissue was reduced to clusters of 20±40 cells. The cells were subcultured on poly-d-lysinecoated petri dishes in serum-free MEM-N3 at a density of 8 £ 104 per dish. Cultures were maintained at 378C in a humidi®ed 6% CO2 atmosphere. In rats, cortical neurons are generated in the ventricular zone between E13 and E19. The newly postmitotic neurons migrate away from ventricular zone and, upon reaching the cortical plate, undergo morphologic differentiation with extension of axons and dendrites. The transition of neural precursor cells to differentiated neurons is also accompanied by antigenic changes. Whereas the precursor cells are immunoreactive for the intermediate ®lament nestin, the postmitotic neurons do not express nestin and instead express markers speci®c for differentiated neurons such as TUJ1. To characterize the developmental stages of the cells, cells were ®xed 4, 24, and 48 h after plating and immunostained with monoclonal antibodies TUJ1 (1:1000, Babco) and nestin (1:400). The morphology of neurons was evaluated microscopically. For each immunocytochemistry experiment, 50 neurons per well, three wells per condition, were randomly selected. Experiments were performed three times (i.e. three separate animals) for each treatment. To avoid counting bias and to ensure that cells were not counted twice, we used the following parameters: neurons whose somata abutted other somata were not counted, as they were considered crowded and hence might not extend neurites optimally; the cells that remained round and did not ¯atten on plastic were not counted; all other cells in the ®eld were counted. A maximum of 10 cells per high-power ®eld (250£) were counted; the ®eld was moved in a ®xed pattern to prevent recounting the same cells and to ensure that cells were sampled equally from the center and edge of the cultures. Immunopositive neurite length was examined at 250£ magni®cation, using a Leitz Ortholux II microscope in conjunction with an Optorix digital video camera, NIH IMAGE 1.61 software. Scales were calibrated by using a microscope scale bar at the same magni®cation. Measurements included percentage of positive cells immunostained by the BrdU, TUJ1, and nestin antibodies (positive cells/ total cells) and neurite length (for TUJ1 immunopositive neurites). (Since an equal number of neurons were sampled under each condition, data are expressed as a percentage of control.) One-way ANOVA was applied to the mean immunopositive cells and neurite lengths to determine differences between control conditions and treatment groups. Data are expressed as mean ^ SD. A level of signi®cance of P , 0:05 was selected. Cells immunopositive for TUJ1 (TUJ1 1) or nestin (nestin 1) were identi®ed at 4, 24, and 48 h after culture. At 4 h after culture, over 80% of the cells were nestin 1 (Figs. 1A,B and 2A) and ,20% of the cells expressed the marker for differentiated neurons (TUJ1 1), indicating that most of these cells had not yet differentiated (Figs. 1C,D and 2B). In contrast, at 24 and 48 h ,20% of cells were nestin 1
Fig. 1. Characterization of E14 cortical cultures. Morphology of E14 cortical cells visualized by phase contrast microscopy at 4 (A,C) and 48 (E,G) h in culture. Immunostaining of cells with a rabbit anti-nestin antibody after 4 (B) and 48 (F) h in cultures and with a mouse anti-TUJ1 antibody at 4 (D) and 48 (H) h in culture.
(Figs. 1E,F and 2A) or glial ®brillary acidic protein (GFAP) 1 (data not shown) and .80% of the cells were TUJ1 1 (Fig. 2B) with extensive neurites, demonstrating the typical morphology of differentiated neurons (Fig. 1G,H). The neurite lengths of these cells were 15 ^ 4.7, 34 ^ 9.1 and 50.8 ^ 11.2 mm (mean ^ SD) after 4, 24 and 48 h, respectively. In the culture system, then, 24 and 48 h after plating, undifferentiated cells progressively disappeared (Fig. 2A), while cells displaying the differentiated neuronal phenotype (Fig. 2B) became the norm.
Fig. 2. Percentage of nestin 1 (A) and TUJ1 1(B) cells after 4, 24 and 48 h in culture, and quantitative effects of T and E (10±100 nM) on the total number of cells (C) and length of neurites (D) of TUJ1 1 cells. *P , 0:05, **P , 0:01, ***P , 0:001.
L. Zhang et al. / Neuroscience Letters 281 (2000) 57±60
To determine the effects of T and E on differentiation, several cultures, plated at the same density, were incubated with or without 10, 50 and 100 nM of T or E. Seventeen beta-estradiol treatment for 24 h increased the length of neurites (Fig. 2D) containing the neuronal marker TUJ1 in a dose-dependent fashion compared with control. By contrast, T (50 and 100 nM) signi®cantly decreased neurite length (Fig. 2D). These results suggest the effects of T and E on neuronal differentiation are remarkably different in early stages of development, with E enhancing and T inhibiting neuronal differentiation. Since T decreased and E increased the number of total cells as well at doses of 50 and 100 nM (Fig. 2C), we examined the effects of E and T on cell proliferation after ®rst observing the extent of proliferation across our culture sampling points. BrdU (Boehringer Mannheim) was diluted in sterile dH2O and added to cultures at a ®nal concentration of 10 mM 4 h before immunostaining. Cells were ®xed with 4% paraformaldehyde in PBS for 15 min and rinsed with PBS, 3 £ 5 min, post®xed in 70% ETOH for 30 min, rinsed with PBS, 2N HCl, for 10 min, rinsed with PBS, and incubated in mouse anti-BrdU (Becton Dickinson) at 48C overnight. Cells were washed extensively after primary antibody incubation, incubated in ¯uorescein-conjugated goat antimouse secondary antibody for 1 h at 258C, washed extensively, and viewed under ¯uorescein epi¯uorescence optics. BrdU labeling of these cultures on the day of plating con®rmed that about 80% of cells proliferated, as determined by nuclear BrdU incorporation (Fig. 3). However, over the 48 h in culture, there was a gradual decline in the fraction of cells that incorporated BrdU and a corresponding increase in the number of cells that developed neurites. This indicates that cell proliferation declines and is replaced with evidence of cellular differentiation after 24 and 48 h culture. To examine whether E and T modulate cell proliferation, the percentages of BrdU positive cells (BrdU 1) in cultures were examined after treatment with 10, 50 and 100 nM E or T for 24 h. We found that neither E nor T signi®cantly changed the number of BrdU positive cells (Fig. 3). This suggests that the changes in the total number of cells produced by either E or T occur not through an alteration in proliferation, but rather through regulation of cell death or survival. Several lines of evidence support the hypothesis that T and E exert their effects through the programmed cell death pathway [1,7]. First, in cortical cultures, T signi®-
Fig. 3. (A) Percentage of BrdU 1 cells at E14 from 4±48 h in culture; (B) percentage of control (BrdU 1 cells) after T and E treatment for 24 h at doses from 10±100 nM. **P , 0:01, ***P , 0:001.
59
cantly increased the expression level of BAX, a protein of the bcl-2 family that promotes apoptosis in developmental CNS (Zhang, unpublished data). Second, administering T to rats between E14 and E18 increased the number of pyknotic cells in the anteroventral periventricular nucleus (AVPv), and administering T to female rats during the ®rst postnatal days stimulated DNA fragmentation in AVPv cells [1]. Third, E upregulated Bcl-2, a protein that primarily suppresses apoptosis in hypothalamic neurons [7]. Finally, an effect of E on Bcl-2 expression has been demonstrated in breast cancer cells [15] and CNS tissues [7]. Our demonstration of opposite effects of E and T on cell number in the absence of effects on proliferation appear consistent with reports that E enhances cell survival while T promotes cell death, although the brain region speci®city of the appearance and direction of these effects is likely. Sex steroids may affect differentiation in the developing cortex through a receptor mediated mechanism [14]. To determine the possible role of AR and ERa in differentiation and proliferation, we examined (1) whether AR and ERa are expressed in the differentiated and undifferentiated neurons of cortex at E14, and (2) whether AR and ERa are colocalized. We ®rst double-labeled neurons with TUJ1 and either AR or ERa antibodies. Next, we double labeled the cells with nestin and either AR or ERa antibodies. Finally, we
Fig. 4. The distribution of AR and ERa -labeled cells in E14 cortical culture after 24 h. (A,C) Morphology of E14 cortical cells visualized by phase contrast microscopy corresponding to (B,D). (B) Cells were double labeled with AR and TUJ1 antibodies. AR 1 (yellow) was expressed in most cells and co-localized with TUJ1 1 (green). (D) Cells were labeled by ERa antibody. ERa was barely detectable in the cultures. (E,F) Immunoblot con®rming the immunoreactivity of ERa and AR antibodies in hypothalamic and cortical tissues from E14 rats. Protein (50 mg) from whole tissues was loaded in wells. Incubation with AR and ERa antibodies revealed the bands of ERa and AR at positions corresponding to reported molecular weights of ERa and AR.
60
L. Zhang et al. / Neuroscience Letters 281 (2000) 57±60
double labeled the cells with AR and ERa . Immunocytochemistry was performed on the cultures by using procedures described above with minor modi®cation (dictated by the different antibodies employed). For detection of the AR and ERa , a polyclonal antibody, generated in rabbit against AR, and a monoclonal antibody, generated in mouse against ERa (Santa Cruz), were used. Western blot analysis using the AR and ERa antibodies con®rmed the immunoreactivity of AR and ERa in the cortex from E14 rat embryos (Fig. 4E,F). We found that AR was expressed in both TUJ1 1 (Fig. 4B) and nestin 1 cells (data not shown) 24 h after plating. However, ER was barely detectable (Fig. 4D). These data suggest that AR but not ER is robustly expressed in both undifferentiated and differentiated cortical neurons at an early developmental stage. Maternal androgens, then, might affect cortical neuronal differentiation at this early stage in a receptor mediated fashion, while any modulatory effects of estradiol in the cortex would, early in development, appear to be receptor independent (or at least not involve ERa ). Consistent with this latter possibility are recent reports of a variety of nuclear estrogen-receptor independent actions of estradiol, including activation of mitogen activated protein (MAP) kinase and antioxidant mediated cell protection [2,16]. The effects of gonadal steroids on neuronal differentiation and survival vary with brain region and developmental state [11,13], and the mechanisms of these effects may similarly vary, suggested by both developmental state speci®c expression of relevant steroid metabolic enzymes (e.g. aromatase and 5 alpha reductase) [11,13] and demonstration of AR and ER dependent morphogenetic effects of gonadal steroids [12]. As we did not attempt to detect the presence of ERb , the involvement of this receptor in cell differentiation can not be ruled out. Further, the requirement in our studies of doses of gonadal steroids greater than those predicted on the basis of receptor Kds suggests that the inference of physiologic relevance cannot be made without caution. Nonetheless, our data are consistent with multiple suggestions that sexual dimorphisms in the brain during development occur in response to different levels of T and E that the brains of male and female animals are exposed to, in¯uencing both neuronal cell death and differentiation. [1] Arai, Y., Murakami, S. and Nishizuka, M., Androgen enhances neuronal degeneration in developing preoptic
[2] [3] [4]
[5]
[6] [7]
[8]
[9] [10]
[11]
[12] [13]
[14] [15]
[16]
area: apoptosis in the anteroventral periventricular nucleus (AVPvN-POA). Horm. Behav., 28 (1994) 313±319. Behl, C., Amyloid beta-protein toxicity and oxidative stress in Alzheimer's disease. Cell Tissue Res., 290 (1997) 471±480. Breedlove, S.M., Sex on the brain. Nature, 389 (1997) 801± 801. Brinton, R.D., Tran, J., Prof®tt, P. and Montoya, M., 17 betaestradiol enhances the outgrowth and survival of neocortical neurons in culture. Neurochem. Res., 22 (1997) 1339± 1351. Brooks, B.P., Merry, D.E., Paulson, H.L., Lieberman, A.P., Kolson, D.L. and Fischbeck, K.H., A cell culture model for androgen effects in motor neurons. J. Neurochem., 70 (1998) 1054±1060. Chowen, J.A., Torres-Aleman, I. and Garcia-Segura, L.M., Trophic effects of estradiol on fetal-rat hypothalamic neurons. Neuroendocrinology, 56 (1992) 895±901. Garcia-Segura, L.M., Cardona-Gomez, P., Naftolin, F. and Chowen, J.A., Estradiol upregulates Bcl-2 expression in adult brain neurons. Neuroendocrinology, 9 (1998) 593± 597. Goldstein, L.A., Kurz, E.M. and Sengelaub, D.R., Androgen regulation of dendritic growth and retraction in the development of a sexually dimorphic spinal nucleus. J. Neurosci., 10 (1990) 935±946. Houtsmuller, E.J., De Jong, F.H., Rowland, D.L. and Slob, A.K., Plasma testosterone in fetal rats and their mothers on day 19 of gestation. Physiol. Behav., 57 (1995) 495±499. Hutchison, J.B., Wozniak, A., Beyer, C., Karolczak, M. and Hutchison, R.E., Steroid metabolizing enzymes in the determination of brain gender. J. Steroid Biochem. Mol. Biol., 69 (1999) 85±96. Lauber, M.E., Ontogeny of 5 alpha-reductase (type 1) messenger ribonucleic acid expression in rat brain: early presence in germinal zones. Endocrinology, 137 (1996) 2718±2730. Lustig, R., Sex hormone modulation of neural development in vitro. Horm. Behav., 28 (1994) 383±395. MacLusky, N.J., Clark, A.S., Naftolin, F. and Goldman-Rakic, P.S., Estrogen formation in the mammalian brain: possible role of aromatase in sexual differentiation of hippocampus and neocortex. Steroids, 50 (1987) 459±474. McEwen, B.S., Steroid hormones: effect on brain development and function. Horm. Res., Suppl., 3 (1992) 1±10. Teixeira, C., Reed, J.C. and Pratt, M.A., Estrogen promotes chemotherapeutic drug resistance by a mechanism involving Bcl-2 proto-oncogene expression in human breast cancer cells. Cancer Res., 55 (1995) 3902±3907. Watters, J.J., Campbell, J.S., Cunningham, M.J., Krebs, E.G. and Dorsa, D.M., Rapid membrane effects of steroids in neuroblastoma cells: effects of estrogen on mitogen activated protein kinase signalling cascade and c-fos immediate early gene transcription. Endocrinology, 138 (1997) 4030±4033.
Testosterone and estrogen affect neuronal differentiation but not proliferation in early embryonic cortex of the rat: the possible roles of androgen and estrogen receptors Lei Zhang a,*, Yoong H. Chang b, Jeffery L. Barker b, Qian Hu b, Li Zhang c, Dragan Maric b, Bing-Sheng Li d, David R. Rubinow a a
Building 36, 2C02 Behavioral Endocrinology Branch, NIMH, NIH, 9000 Rockville Pike, Bethesda, MD 20892, USA b Laboratory of Neurophysiology, NINDS, NIH, Bethesda, MD 20892, USA c Synaptic Function Unit, NINDS, NIH, Bethesda, MD 20892, USA d Laboratory of Neurochemistry, NINDS, NIH, Bethesda, MD 20892, USA Received 3 December 1999; received in revised form 10 January 2000; accepted 11 January 2000
Abstract We examined the effect of testosterone (T) and 17 beta-estradiol (E) on differentiation and proliferation of cultured neurons from the cortex of 14-day-rat embryos (E14) using immunocytochemistry. We found that the cultures receiving E had signi®cantly more neurons with longer neurites than the control cultures, while both fewer and less differentiated neurons were seen after 24 h of incubation with T. However, neither T nor E changed the number of cells positive for BrdU, a proliferation marker. We also found that the androgen receptor (AR) was markedly expressed in the neurons, whereas the expression of estrogen receptor (ERa ) was barely detectable. These results suggest that E and T differ in effect on differentiation, while neither affect proliferation in early developmental cortex. Furthermore, since the AR is expressed in the cortical neurons by E14, the inhibitory effect of T on differentiation may be receptor-mediated, while the stimulatory effects of estrogen in the cortex do not appear to involve nuclear ERa at this developmental stage. Published by Elsevier Science Ireland Ltd. Keywords: Testosterone; Estrogen; Sexual dimorphism; Cortex; Differentiation; Proliferation
Sex steroids, including testosterone (T) and estradiol (E), help direct the sexual differentiation and cellular proliferation of the brain [6,14]. Estrogen stimulates neural differentiation and modulates neural survival both in vitro and in vivo [4,6]. Similarly, androgens such as T are known to affect cell growth, differentiation, morphology, and survival [5,8]. Further, at least some of the effects of T on neuronal morphology and differentiation are mediated by its conversion to E [10]. The mechanisms by which T and E regulate neuronal differentiation are essentially unknown. If the sexually dimorphic nature of the brain is determined primarily by sex-related differences in sex hormone levels (although see Breedlove [3] and McEwen [14] for contribution of behavioral and genetic factors), one should be able to
* Corresponding author. Tel.: 11-301-402-1398; fax: 11-301402-1565. E-mail address: [email protected] (L. Zhang)
demonstrate a differential effect of T and E on cellular differentiation. To test this hypothesis, we ®rst developed an embryonic culture system in which we could observe both neuronal proliferation and differentiation. Second, we examined the effects of T and E on proliferation and differentiation, measuring cell number, bromodeoxyuridine (BrdU) labeling (a proliferation marker), and neurite length in cells labeled with the class III neuron-speci®c b -tubulin protein (TUJ1). Finally, we examined the expression and distribution of androgen receptors (AR) and estrogen receptors (ER) in the cultures by immunocytochemistry to determine the possible mechanism for any observed responses to the sex steroids. Cultured neurons were derived from the cerebral cortex of Sprague±Dawley rat embryos (Taconic) on embryonic day 14 (E14), the time when the ®rst postmitotic neurons of the cortex are beginning to differentiate, cell proliferation is predominant, and sex-related differences in the levels of T in the circulation have not appeared [9]. The cerebral hemisphere was dissected and gently triturated with
0304-3940/00/$ - see front matter. Published by Elsevier Science Ireland Ltd. PII: S03 04 - 394 0( 9 9) 00 94 2- 8
58
L. Zhang et al. / Neuroscience Letters 281 (2000) 57±60
a pipette, so that most of the tissue was reduced to clusters of 20±40 cells. The cells were subcultured on poly-d-lysinecoated petri dishes in serum-free MEM-N3 at a density of 8 £ 104 per dish. Cultures were maintained at 378C in a humidi®ed 6% CO2 atmosphere. In rats, cortical neurons are generated in the ventricular zone between E13 and E19. The newly postmitotic neurons migrate away from ventricular zone and, upon reaching the cortical plate, undergo morphologic differentiation with extension of axons and dendrites. The transition of neural precursor cells to differentiated neurons is also accompanied by antigenic changes. Whereas the precursor cells are immunoreactive for the intermediate ®lament nestin, the postmitotic neurons do not express nestin and instead express markers speci®c for differentiated neurons such as TUJ1. To characterize the developmental stages of the cells, cells were ®xed 4, 24, and 48 h after plating and immunostained with monoclonal antibodies TUJ1 (1:1000, Babco) and nestin (1:400). The morphology of neurons was evaluated microscopically. For each immunocytochemistry experiment, 50 neurons per well, three wells per condition, were randomly selected. Experiments were performed three times (i.e. three separate animals) for each treatment. To avoid counting bias and to ensure that cells were not counted twice, we used the following parameters: neurons whose somata abutted other somata were not counted, as they were considered crowded and hence might not extend neurites optimally; the cells that remained round and did not ¯atten on plastic were not counted; all other cells in the ®eld were counted. A maximum of 10 cells per high-power ®eld (250£) were counted; the ®eld was moved in a ®xed pattern to prevent recounting the same cells and to ensure that cells were sampled equally from the center and edge of the cultures. Immunopositive neurite length was examined at 250£ magni®cation, using a Leitz Ortholux II microscope in conjunction with an Optorix digital video camera, NIH IMAGE 1.61 software. Scales were calibrated by using a microscope scale bar at the same magni®cation. Measurements included percentage of positive cells immunostained by the BrdU, TUJ1, and nestin antibodies (positive cells/ total cells) and neurite length (for TUJ1 immunopositive neurites). (Since an equal number of neurons were sampled under each condition, data are expressed as a percentage of control.) One-way ANOVA was applied to the mean immunopositive cells and neurite lengths to determine differences between control conditions and treatment groups. Data are expressed as mean ^ SD. A level of signi®cance of P , 0:05 was selected. Cells immunopositive for TUJ1 (TUJ1 1) or nestin (nestin 1) were identi®ed at 4, 24, and 48 h after culture. At 4 h after culture, over 80% of the cells were nestin 1 (Figs. 1A,B and 2A) and ,20% of the cells expressed the marker for differentiated neurons (TUJ1 1), indicating that most of these cells had not yet differentiated (Figs. 1C,D and 2B). In contrast, at 24 and 48 h ,20% of cells were nestin 1
Fig. 1. Characterization of E14 cortical cultures. Morphology of E14 cortical cells visualized by phase contrast microscopy at 4 (A,C) and 48 (E,G) h in culture. Immunostaining of cells with a rabbit anti-nestin antibody after 4 (B) and 48 (F) h in cultures and with a mouse anti-TUJ1 antibody at 4 (D) and 48 (H) h in culture.
(Figs. 1E,F and 2A) or glial ®brillary acidic protein (GFAP) 1 (data not shown) and .80% of the cells were TUJ1 1 (Fig. 2B) with extensive neurites, demonstrating the typical morphology of differentiated neurons (Fig. 1G,H). The neurite lengths of these cells were 15 ^ 4.7, 34 ^ 9.1 and 50.8 ^ 11.2 mm (mean ^ SD) after 4, 24 and 48 h, respectively. In the culture system, then, 24 and 48 h after plating, undifferentiated cells progressively disappeared (Fig. 2A), while cells displaying the differentiated neuronal phenotype (Fig. 2B) became the norm.
Fig. 2. Percentage of nestin 1 (A) and TUJ1 1(B) cells after 4, 24 and 48 h in culture, and quantitative effects of T and E (10±100 nM) on the total number of cells (C) and length of neurites (D) of TUJ1 1 cells. *P , 0:05, **P , 0:01, ***P , 0:001.
L. Zhang et al. / Neuroscience Letters 281 (2000) 57±60
To determine the effects of T and E on differentiation, several cultures, plated at the same density, were incubated with or without 10, 50 and 100 nM of T or E. Seventeen beta-estradiol treatment for 24 h increased the length of neurites (Fig. 2D) containing the neuronal marker TUJ1 in a dose-dependent fashion compared with control. By contrast, T (50 and 100 nM) signi®cantly decreased neurite length (Fig. 2D). These results suggest the effects of T and E on neuronal differentiation are remarkably different in early stages of development, with E enhancing and T inhibiting neuronal differentiation. Since T decreased and E increased the number of total cells as well at doses of 50 and 100 nM (Fig. 2C), we examined the effects of E and T on cell proliferation after ®rst observing the extent of proliferation across our culture sampling points. BrdU (Boehringer Mannheim) was diluted in sterile dH2O and added to cultures at a ®nal concentration of 10 mM 4 h before immunostaining. Cells were ®xed with 4% paraformaldehyde in PBS for 15 min and rinsed with PBS, 3 £ 5 min, post®xed in 70% ETOH for 30 min, rinsed with PBS, 2N HCl, for 10 min, rinsed with PBS, and incubated in mouse anti-BrdU (Becton Dickinson) at 48C overnight. Cells were washed extensively after primary antibody incubation, incubated in ¯uorescein-conjugated goat antimouse secondary antibody for 1 h at 258C, washed extensively, and viewed under ¯uorescein epi¯uorescence optics. BrdU labeling of these cultures on the day of plating con®rmed that about 80% of cells proliferated, as determined by nuclear BrdU incorporation (Fig. 3). However, over the 48 h in culture, there was a gradual decline in the fraction of cells that incorporated BrdU and a corresponding increase in the number of cells that developed neurites. This indicates that cell proliferation declines and is replaced with evidence of cellular differentiation after 24 and 48 h culture. To examine whether E and T modulate cell proliferation, the percentages of BrdU positive cells (BrdU 1) in cultures were examined after treatment with 10, 50 and 100 nM E or T for 24 h. We found that neither E nor T signi®cantly changed the number of BrdU positive cells (Fig. 3). This suggests that the changes in the total number of cells produced by either E or T occur not through an alteration in proliferation, but rather through regulation of cell death or survival. Several lines of evidence support the hypothesis that T and E exert their effects through the programmed cell death pathway [1,7]. First, in cortical cultures, T signi®-
Fig. 3. (A) Percentage of BrdU 1 cells at E14 from 4±48 h in culture; (B) percentage of control (BrdU 1 cells) after T and E treatment for 24 h at doses from 10±100 nM. **P , 0:01, ***P , 0:001.
59
cantly increased the expression level of BAX, a protein of the bcl-2 family that promotes apoptosis in developmental CNS (Zhang, unpublished data). Second, administering T to rats between E14 and E18 increased the number of pyknotic cells in the anteroventral periventricular nucleus (AVPv), and administering T to female rats during the ®rst postnatal days stimulated DNA fragmentation in AVPv cells [1]. Third, E upregulated Bcl-2, a protein that primarily suppresses apoptosis in hypothalamic neurons [7]. Finally, an effect of E on Bcl-2 expression has been demonstrated in breast cancer cells [15] and CNS tissues [7]. Our demonstration of opposite effects of E and T on cell number in the absence of effects on proliferation appear consistent with reports that E enhances cell survival while T promotes cell death, although the brain region speci®city of the appearance and direction of these effects is likely. Sex steroids may affect differentiation in the developing cortex through a receptor mediated mechanism [14]. To determine the possible role of AR and ERa in differentiation and proliferation, we examined (1) whether AR and ERa are expressed in the differentiated and undifferentiated neurons of cortex at E14, and (2) whether AR and ERa are colocalized. We ®rst double-labeled neurons with TUJ1 and either AR or ERa antibodies. Next, we double labeled the cells with nestin and either AR or ERa antibodies. Finally, we
Fig. 4. The distribution of AR and ERa -labeled cells in E14 cortical culture after 24 h. (A,C) Morphology of E14 cortical cells visualized by phase contrast microscopy corresponding to (B,D). (B) Cells were double labeled with AR and TUJ1 antibodies. AR 1 (yellow) was expressed in most cells and co-localized with TUJ1 1 (green). (D) Cells were labeled by ERa antibody. ERa was barely detectable in the cultures. (E,F) Immunoblot con®rming the immunoreactivity of ERa and AR antibodies in hypothalamic and cortical tissues from E14 rats. Protein (50 mg) from whole tissues was loaded in wells. Incubation with AR and ERa antibodies revealed the bands of ERa and AR at positions corresponding to reported molecular weights of ERa and AR.
60
L. Zhang et al. / Neuroscience Letters 281 (2000) 57±60
double labeled the cells with AR and ERa . Immunocytochemistry was performed on the cultures by using procedures described above with minor modi®cation (dictated by the different antibodies employed). For detection of the AR and ERa , a polyclonal antibody, generated in rabbit against AR, and a monoclonal antibody, generated in mouse against ERa (Santa Cruz), were used. Western blot analysis using the AR and ERa antibodies con®rmed the immunoreactivity of AR and ERa in the cortex from E14 rat embryos (Fig. 4E,F). We found that AR was expressed in both TUJ1 1 (Fig. 4B) and nestin 1 cells (data not shown) 24 h after plating. However, ER was barely detectable (Fig. 4D). These data suggest that AR but not ER is robustly expressed in both undifferentiated and differentiated cortical neurons at an early developmental stage. Maternal androgens, then, might affect cortical neuronal differentiation at this early stage in a receptor mediated fashion, while any modulatory effects of estradiol in the cortex would, early in development, appear to be receptor independent (or at least not involve ERa ). Consistent with this latter possibility are recent reports of a variety of nuclear estrogen-receptor independent actions of estradiol, including activation of mitogen activated protein (MAP) kinase and antioxidant mediated cell protection [2,16]. The effects of gonadal steroids on neuronal differentiation and survival vary with brain region and developmental state [11,13], and the mechanisms of these effects may similarly vary, suggested by both developmental state speci®c expression of relevant steroid metabolic enzymes (e.g. aromatase and 5 alpha reductase) [11,13] and demonstration of AR and ER dependent morphogenetic effects of gonadal steroids [12]. As we did not attempt to detect the presence of ERb , the involvement of this receptor in cell differentiation can not be ruled out. Further, the requirement in our studies of doses of gonadal steroids greater than those predicted on the basis of receptor Kds suggests that the inference of physiologic relevance cannot be made without caution. Nonetheless, our data are consistent with multiple suggestions that sexual dimorphisms in the brain during development occur in response to different levels of T and E that the brains of male and female animals are exposed to, in¯uencing both neuronal cell death and differentiation. [1] Arai, Y., Murakami, S. and Nishizuka, M., Androgen enhances neuronal degeneration in developing preoptic
[2] [3] [4]
[5]
[6] [7]
[8]
[9] [10]
[11]
[12] [13]
[14] [15]
[16]
area: apoptosis in the anteroventral periventricular nucleus (AVPvN-POA). Horm. Behav., 28 (1994) 313±319. Behl, C., Amyloid beta-protein toxicity and oxidative stress in Alzheimer's disease. Cell Tissue Res., 290 (1997) 471±480. Breedlove, S.M., Sex on the brain. Nature, 389 (1997) 801± 801. Brinton, R.D., Tran, J., Prof®tt, P. and Montoya, M., 17 betaestradiol enhances the outgrowth and survival of neocortical neurons in culture. Neurochem. Res., 22 (1997) 1339± 1351. Brooks, B.P., Merry, D.E., Paulson, H.L., Lieberman, A.P., Kolson, D.L. and Fischbeck, K.H., A cell culture model for androgen effects in motor neurons. J. Neurochem., 70 (1998) 1054±1060. Chowen, J.A., Torres-Aleman, I. and Garcia-Segura, L.M., Trophic effects of estradiol on fetal-rat hypothalamic neurons. Neuroendocrinology, 56 (1992) 895±901. Garcia-Segura, L.M., Cardona-Gomez, P., Naftolin, F. and Chowen, J.A., Estradiol upregulates Bcl-2 expression in adult brain neurons. Neuroendocrinology, 9 (1998) 593± 597. Goldstein, L.A., Kurz, E.M. and Sengelaub, D.R., Androgen regulation of dendritic growth and retraction in the development of a sexually dimorphic spinal nucleus. J. Neurosci., 10 (1990) 935±946. Houtsmuller, E.J., De Jong, F.H., Rowland, D.L. and Slob, A.K., Plasma testosterone in fetal rats and their mothers on day 19 of gestation. Physiol. Behav., 57 (1995) 495±499. Hutchison, J.B., Wozniak, A., Beyer, C., Karolczak, M. and Hutchison, R.E., Steroid metabolizing enzymes in the determination of brain gender. J. Steroid Biochem. Mol. Biol., 69 (1999) 85±96. Lauber, M.E., Ontogeny of 5 alpha-reductase (type 1) messenger ribonucleic acid expression in rat brain: early presence in germinal zones. Endocrinology, 137 (1996) 2718±2730. Lustig, R., Sex hormone modulation of neural development in vitro. Horm. Behav., 28 (1994) 383±395. MacLusky, N.J., Clark, A.S., Naftolin, F. and Goldman-Rakic, P.S., Estrogen formation in the mammalian brain: possible role of aromatase in sexual differentiation of hippocampus and neocortex. Steroids, 50 (1987) 459±474. McEwen, B.S., Steroid hormones: effect on brain development and function. Horm. Res., Suppl., 3 (1992) 1±10. Teixeira, C., Reed, J.C. and Pratt, M.A., Estrogen promotes chemotherapeutic drug resistance by a mechanism involving Bcl-2 proto-oncogene expression in human breast cancer cells. Cancer Res., 55 (1995) 3902±3907. Watters, J.J., Campbell, J.S., Cunningham, M.J., Krebs, E.G. and Dorsa, D.M., Rapid membrane effects of steroids in neuroblastoma cells: effects of estrogen on mitogen activated protein kinase signalling cascade and c-fos immediate early gene transcription. Endocrinology, 138 (1997) 4030±4033.