Signaling pathways involved with serotonin1A agonist-mediated neuroprotection against ethanol-induced apoptosis of fetal rhombencephalic neurons

Developmental Brain Research 159 (2005) 18 – 28 www.elsevier.com/locate/devbrainres

Research Report

Signaling pathways involved with serotonin1A agonist-mediated neuroprotection against ethanol-induced apoptosis of fetal rhombencephalic neurons Mary Drusea,b,c,*, Nuzhath F. Tajuddina,b, Roberta A. Gillespiea,b, Phong Leb a Division of Molecular and Cellular Biochemistry, Loyola University Stritch School of Medicine, 2160 S. First Avenue, Maywood, IL 60153, USA Department of Cell Biology, Neurobiology and Anatomy, Loyola University Stritch School of Medicine, 2160 S. First Avenue, Maywood, IL 60153, USA c The Alcohol Research Program, Loyola University Stritch School of Medicine, 2160 S. First Avenue, Maywood, IL 60153, USA

b

Accepted 8 June 2005 Available online 2 August 2005

Abstract Previously, this laboratory demonstrated that developing serotonin (5-HT) neurons and other fetal rhombencephalic neurons are reduced by in vivo and in vitro exposure to ethanol, effects that are related to ethanol’s augmentation of apoptosis. We also found that 5-HT1A agonists diminished the ethanol-associated reduction of 5-HT neurons and other fetal rhombencephalic neurons by attenuating the proapoptotic effects of ethanol. Presently, we investigated the hypothesis that the protective/anti-apoptotic effects of a 5-HT1A agonist on fetal rhombencephalic neurons are mediated by activation of the phosphatidylinositol 3V kinase (PI-3K) and/or the mitogen-activated protein kinase kinase (MAPKK) pathway. Apoptotic and non-apoptotic fetal rhombencephalic neurons were quantitated in primary cultures that were treated with 50 mM ethanol and with 100 nM of a 5-HT1A agonist such as 8-OH-DPAT [8-hydroxy 2-(di-n-propylamino)tetralin], ipsapirone, or buspirone. Analysis of neurons stained with Hoechst 33342 demonstrated the anti-apoptotic effects of 5-HT1A agonists and implicated the involvement of the PI-3K pathway and possibly the MAPKK pathway with the protective effects of these drugs. The protective effects were blocked by a 5-HT1A antagonist (WAY 100635), an inhibitor of PI-3K (LY294002), and an inhibitor of MAPKK (PD98059). Western blot analyses showed that ethanol treatment reduces basal pAkt levels. These analyses also provide support for the involvement of the PI-3K pathway; ipsapirone stimulated the phosphorylation of Akt in control and ethanol-treated neurons, and these effects were antagonized by LY294002. D 2005 Elsevier B.V. All rights reserved. Theme: Development and regeneration Topic: Neuronal death Keywords: Serotonin1A receptor; Apoptosis; Neuroprotection; Akt; MAPK

1. Introduction A severe and devastating consequence of in utero ethanol exposure in humans is CNS dysfunction. Examples of such CNS dysfunction include reduced intelligence and learning * Corresponding author. Department of Cell Biology, Neurobiology and Anatomy, Loyola University Stritch School of Medicine, 2160 S. First Avenue, Maywood, IL 60153, USA. Fax: +1 708 216 8523. E-mail address: [email protected] (M. Druse-Manteuffel). 0165-3806/$ - see front matter D 2005 Elsevier B.V. All rights reserved. doi:10.1016/j.devbrainres.2005.06.015

abnormalities. Some children exposed to ethanol in utero also exhibit impaired linguistic ability as well as abnormal psychomotor and mental function [14,15,35,45]. In addition, offspring of heavy prenatal alcohol drinkers exhibit mental illness [25] and abnormalities in psychosocial function [54]. Although multiple mechanisms are likely to underlie the damaging effects of ethanol on the fetal brain, one important mechanism involves the high susceptibility of developing neurons to ethanol-associated cell death. Previous reports emphasize that ethanol reduces the number of developing

M. Druse et al. / Developmental Brain Research 159 (2005) 18 – 28

cerebellar granule and Purkinje neurons [9,27,43,44], hippocampal neurons [46,47], and cortical neurons [34]. Studies from our own and another laboratory show that developing serotonin (5-HT) neurons are also highly vulnerable to the damaging effects of ethanol. Following in utero ethanol exposure, there is a significant deficit of 5HT neurons, 5-HT content, and 5-HT reuptake sites [52,56,58,59,64,65]. The damaging effects of ethanol on 5-HT neurons and the associated reduction of fetal brain 5-HT [52,58] are likely to augment ethanol’s effects on the developing brain. A normal concentration of fetal 5-HT, an essential neurotrophic factor, is required to promote the development of 5-HT neurons, targets of 5-HT neurons, and other CNS neurons [8,37,39–41,63]. The developmental effects of 5-HT include the modulation of growth cone elongation and of neurite outgrowth; 5-HT also exerts an autocrine effect on 5-HT synthesis [17,29]. Many of the trophic effects of 5-HT are mediated both by neuronal and astrocyte 5-HT1A receptors [8,62]. Previous reports from this and another laboratory support the involvement of an early 5-HT deficit with the ethanolassociated impaired development of the 5-HT system. In fact, early treatment with a 5-HT1A agonist such as buspirone or ipsapirone blocks most of the ethanolassociated abnormalities in the developing 5-HT system. Treatment of ethanol-fed pregnant rats with a 5-HT1A agonist (e.g., buspirone or ipsapirone) prevents the reduced density of 5-HT neurons in the dorsal and median raphe. This treatment also blocks the ethanol-associated decrease in the concentrations of 5-HT and 5-HT reuptake sites in most brain areas examined [37,58 – 60]. Recently, this laboratory reported that ethanol impairs developing 5-HT neurons by augmenting apoptosis in these and other fetal rhombencephalic neurons [21]. Other laboratories detect ethanol-induced apoptosis in developing neural crest cells [11,22,42] as well as in cortical neurons [13], cerebellar granule neurons [48,49], and forebrain neurons [33]. Importantly, this laboratory found that the 5HT1A agonist ipsapirone attenuated ethanol-associated apoptosis of 5-HT and other fetal rhombencephalic neurons [21]. Traditionally, activation of the 5-HT1A receptor has been linked to the inhibition of adenylyl cyclase [18,31], increased K+ conductance [6], and decreased Ca2+ conductance [50]. Although the specific cellular signaling mechanism(s) by which 5-HT1A agonists prevent apoptosis in vivo or in primary neuronal cultures has not been definitively established, investigations using transfected cell lines [26,53] suggest that phosphatidylinositol 3Vkinase (PI-3K) and/or mitogen-activated protein kinase kinase (MAPKK) pathways might be involved. In the present study, we investigated the hypothesis that 5-HT1A agonists activate PI-3K and the subsequent phosphorylation of Akt (pAkt) in fetal rhombencephalic neurons and that activation of PI-3K contributes to the antiapoptotic effects of 5-HT1A agonists on ethanol-associated apoptosis in these neurons. We also investigated the

19

hypothesis that 5-HT1A agonists activate MAPKK and the downstream phosphorylation of MAPK/ERK-1 (p42/ p44 MAPK/extracellular signal-regulated kinase-1) in fetal rhombencephalic neurons and that activation of MAPKK contributes to the anti-apoptotic effects of 5-HT1A agonists on ethanol-associated apoptosis in these neurons. A 5HT1A receptor antagonist and inhibitors of the PI-3K and MAPKK enzymes were also used to elucidate the signaling mechanism(s) underlying the protective effects of a 5-HT1A agonist on ethanol-treated fetal rhombencephalic neurons.

2. Materials and methods 2.1. Tissue dissection and dissociation of cells This laboratory previously described the methods used for tissue dissection and for preparation of primary cultures of fetal rhombencephalic neurons [24]. In order to collect tissue that contains the developing serotonergic raphe neurons, the rhombencephalon was removed from embryos taken from timed-pregnancy Sprague – Dawley rats at G14, where G = 0 corresponds to the day of insemination. The G14 rhombencephalon is used because the cell bodies of 5HT immunoreactive neurons are localized in this brain region at G14 [38]. All animal care and use procedures were reviewed and approved by the Institutional Animal Care and Use Committee at Loyola University Chicago, Stritch School of Medicine. Dissected tissue from 2 to 3 litters per experiment was mechanically disaggregated using 230 and 130 Am Nitex bags (Sefar America, HR Williams Division, Kansas City, MO) as described previously and the suspension was collected. 2.2. Neuronal cell cultures On the day of dissection, i.e., day 1 in vitro (DIV1), cells were plated onto poly-d-lysine-coated four-chambered slides (Nalge and Nunc, Naperville, IL) at a density of 250,000 – 300,000 cells/chamber (1.8 cm2) for staining with Hoechst 33342. Alternatively, cells were plated onto polyd-lysine-coated plates (Corning, Corning, New York) at a density of 8  106 to 10  106 cells/plate (55 cm2) for Western blot studies. We used a neuron-specific chemically defined media (CDM) that was supplemented with 0.25% fetal bovine serum (FBS), because our laboratory previously used this media to demonstrate both the damaging effects of 50 mM ethanol and the neuroprotective effects of ipsapirone [21]. The CDM is a Dulbecco’s minimal essential media/ F12 (DMEM/F12) media, containing hydrocortisone-21 sulfate, Basal Medium Eagle Vitamin Solution, antibacterial agent gentamicin sulfate [30], and B27 serum-free medium supplement [10]. Cells were grown in the neuron-specific media and then treated with 0.4 AM cytosine arabinoside (Sigma-Aldrich, St. Louis, MO) (ara-C) after 24 h and at subsequent media changes in order to arrest astrocyte

20

M. Druse et al. / Developmental Brain Research 159 (2005) 18 – 28

growth and proliferation. Cells were grown in control media (no ethanol) 5 days (DIV1 to DIV6) with media changes on alternate days; either 0 mM ethanol (control group) or 50 mM ethanol (ethanol group) was present in the cultures for the last 24 h (DIV5 to DIV6). In studies of Hoechst-stained apoptotic neurons, other agents were also present during the last 24 h: 100 nM of a 5-HT1A agonist [ipsapirone or 8-hydroxy 2-(di-n-propylamino)tetralin (8-OH-DPAT)], 1 AM WAY 100635 (a 5HT1A receptor antagonist), 10 AM LY294002 (a PI-3K inhibitor), or 10 AM PD98059 (a MAPKK inhibitor). Media were changed after 24 h, on alternate days, and at the time of treatment. A concentration of 100 nM ipsapirone was used because this concentration promotes the development of 5-HT neurons [62] and prevents apoptosis in fetal rhombencephalic neurons [21]. The 10AM concentrations of LY294002 and PD98059 were used because this laboratory finds that these inhibitor concentrations effectively (>90%) reduce pAkt and p42/p44 MAPK in unstimulated cultures; we also find that these concentrations do not have a marked effect on the other signaling pathway under investigation. For studies using Western blot analyses, we replaced media with a serum-free CDM 1 h prior to the addition of a 5-HT1A agonist. A mixture of phosphatase inhibitors (200 AM sodium orthovanadate and 10 mM sodium fluoride) was also added at that time. Ten minutes prior to the termination of the experiment, the 5-HT1A agonist ipsapirone was added at a concentration of 100 nM. The receptor specificity of the effects of this 5-HT1A agonist was determined using 100 AM of the 5-HT1A antagonist WAY 100635. The potential involvement of the PI-3K and MAPKK pathways was determined using, respectively, 10 AM of LY294002 or 10 AM of PD98059. 2.3. Ethanol chamber system As described in detail in an earlier paper from this laboratory [23], the ethanol concentration in the media was maintained at 85% of that initially established in the chamber system. Control or ethanol-containing culture dishes were placed into large polypropylene containers, which contained 200 ml of an ethanol– water mixture of equivalent molarity to that of ethanol-treated cell cultures. Media were replaced on alternate days, and the ethanolcontaining water baths were replenished every 2 days or whenever the large containers were open for more than 20 min. 2.4. Identification of live cells and those with fragmented apoptotic nuclei using Hoechst 33342 In this study, we identified apoptotic cells that were stained with Hoechst 33342 according to an established method [7]. Previously, we reported that Hoechst 33342, which visualizes fragmented nuclei, and TUNEL (terminal

deoxynucleotidyl transferase-mediated dUTP – biotin nick end-labeling), which labels fragmented DNA in apoptotic cells, identify the identical population of apoptotic fetal rhombencephalic neurons in control and ethanol-treated cultures [21]. In the present study, cultured cells were fixed in 4% formaldehyde in phosphate-buffered saline (PBS), rinsed three times with PBS, and incubated with a 1:500 dilution of stock solution of Hoechst 33342 (1.2 mg/ml) in PBS in a humidified chamber at 37 -C for 15 min. Slides were rinsed five times with PBS. A Nikon Microphot fluorescence microscope was used in the ultraviolet range to view Hoechst-stained cells. Images were captured using a 40 objective and analyzed at a higher, computer-enhanced magnification (100). As described previously [21], neurons that were identified as apoptotic appeared to be intact and contained fragmented apoptotic nuclei. If necrotic cells had been present, they would be expected to exhibit signs of swelling, and their chromatin would be diffuse or finely clumped. Neurons that were identified as alive/nonapoptotic exhibited an intact cellular morphology and lacked fragmented nuclei. As shown in our previous study [21], the presence of ara-C reduced the percentage of cells that were immunopositive for the astrocyte marker GFAP (glial fibrillary acidic protein) to less than 5%; any residual glia had the appearance of small dots. Extracellular debris, fragmented cells, and Fdots_, whether of neuronal or glial origin, were not included in the cell counts. 2.5. Western blot analyses At the end of the experiments, media were removed, cells were rinsed with 0.01 M PBS, pH 7.6, and lysis buffer was added. Cells were scraped into the lysis buffer that contained protease inhibitors (leupeptin, aprotinin, pepstatin A, phenylmethylsulfonyl fluoride) and then sonicated. Denatured and reduced proteins were separated by SDS–PAGE on 10% polyacrylamide gels. Molecular weight markers (Bio-Rad, Hercules, CA) were also included on each gel. Following electrophoresis, proteins were transferred to nitrocellulose membranes, using 5% milk in Tris-buffered saline containing 0.05% Tween-20 as a blocking agent. Blots were probed either with a primary antibody against the phosphorylated form of Akt (pAkt) or against p42/p44 MAPK (Cell Signaling, Beverly, MA). As described previously [20], we used a secondary antibody that was conjugated to horse radish peroxidase and ECL (excitation of chemiluminescence) detection. Care was taken to overlay each film on its paired blot to insure that bands of the appropriate molecular weights were identified by the antibodies to pAkt and p42/ p44MAPK. The relative optical densities of the bands that corresponded to either pAkt or p42 and p44MAPK on the autoradiograms were determined according to the directions provided with the NIH Macintosh-based image analysis program: Image. The densities of the major band, p42 MAPK, and the minor band, p44 MAPK, were combined. Several steps were taken to insure that a comparable amount

M. Druse et al. / Developmental Brain Research 159 (2005) 18 – 28

of protein was loaded onto each gel and transferred to each blot. (1) An identical amount of protein, e.g., 20 Ag, was loaded onto each lane of a gel. (2) Efficient and comparable transfer of proteins to blots was determined by examining any residual proteins on Coomassie blue-stained gel. (3) The density of actin on Coomassie blue-stained gels was evaluated as a loading control. 2.6. Statistical analyses Treatment effects were determined using ANOVA and a post hoc Tukey’s test (P < 0.05). Although all statistical analyses were performed using actual experimental data,

21

treatment effects are typically depicted as the percentage of the means in control cultures.

3. Results 3.1. Ethanol reduces survival and promotes apoptosis of fetal rhombencephalic neurons in vitro, and a 5-HT1A agonist such as ipsapirone or 8-OH-DPAT prevents the pro-apoptotic effects of 50 mM ethanol (Figs. 1 and 2) Fig. 1 depicts fetal rhombencephalic neurons from cultures that were maintained under control conditions or

Fig. 1. Hoechst 33342-stained fetal rhombencephalic neurons that were cultured in the presence of ethanol, a 5-HT1A agonist, LY294002, WAY 100635, and/or PD98059. Fetal rhombencephalic neurons stained with Hoechst 33342 from cultures that were maintained under control conditions (A, C, E, G, I) or that were treated for the last 24 h (DIV5 to DIV6) with the 50-mM ethanol (B, D, F, H, J) and/or the following agents: a 5-HT1A agonist such as ipsapirone or 8-OH-DPAT (C – J), LY294002 (E, F), WAY 100635 (G, H), PD98059 (I, J). Hoechst-stained living and apoptotic neurons were identified as described in the methods section. A higher magnification was used to identify neurons that exhibit the characteristics of apoptotic cells; these apoptotic cells are shown by the arrows.

22

M. Druse et al. / Developmental Brain Research 159 (2005) 18 – 28

that were treated for the last 24 h (DIV5 to DIV6) with ethanol, a 5-HT1A agonist such as ipsapirone or 8-OHDPAT, LY294002, WAY 100635, or PD98059. These neurons were stained with Hoechst 33342. Using data obtained from multiple experiments, the percentage of surviving (non-apoptotic) and apoptotic neurons was determined [% surviving neurons = (No. of non-apoptotic neurons) / (No. of non-apoptotic neurons + No. of apoptotic neurons)], and these data are summarized in Figs. 2A and 2B. The graphic depiction of the results from the same experiments is divided into groups that were cultured in the presence (2A) and absence (2B) of a 5-HT1A agonist; data from basal/unstimulated control and ethanol-exposed cultures are repeated in both figures. The data presented in these figures demonstrate that a 24-h treatment of fetal neurons with 50 mM ethanol (¨230 mg/dl) significantly increases apoptosis and consequently reduces the percentage

of surviving neurons [F(1,102) = 47.1, P < 0.0001]. Post hoc analyses confirmed that the percentage of surviving (non-apoptotic) fetal rhombencephalic neurons was significantly reduced by ethanol (P < 0.01); the percentage of surviving/non-apoptotic neurons in control cultures (C) was 77.8 T 1.0%, while that in ethanol-treated cultures (E) was 56.2 T 1.4%. However, in comparison with ethanol-treated neurons, there is an increased percentage of non-apoptotic/ surviving neurons in cultures that were co-treated with ethanol and a 5-HT1A agonist (E1A), and the latter values are comparable to those in control cultures. Nonetheless, 5HT1A agonist treatment (C1A) does not increase the percentage of surviving neurons in control cultures. Data presented in Figs. 2A and 2B were combined for the two 5HT1A agonists used in these studies, ipsapirone and 8-OHDPAT, because these drugs exerted comparable effects on cultured neurons; the percentages of surviving neurons in

Fig. 2. Ethanol reduces survival and promotes apoptosis of fetal rhombencephalic neurons in vitro, and a 5-HT1A agonist such as ipsapirone or 8-OH-DPAT prevents the pro-apoptotic effects of 50 mM ethanol. (A, B) Graphic depictions of the percentage of surviving/non-apoptotic fetal rhombencephalic neurons. These figures include data obtained from groups treated with the following conditions: control, no ethanol (C); ethanol (E); a 5-HT1A agonist (either 8-OHDPAT or ipsapirone, 1A); LY294002 (LY), an inhibitor of PI-3K; WAY 100635 (W), a 5-HT1A antagonist; or PD98059 (PD), an inhibitor of MAPKK. Statistical differences (P < 0.05, P < 0.01) are noted for the following comparisons: (a, aa) control (C) relative to all other groups; (b, bb) control + 5-HT1A agonist that received no additional treatment (C1A) relative to the control + 5-HT1A agonist groups that were co-treated with either LY, PD, or W; (cc) ethanol + 5-HT1A agonist-treated group (E1A) that received no further treatment relative to ethanol (E) and to the ethanol + 5-HT1A agonist-treated groups that were cotreated with either LY, PD, or W. The data that are presented in (A, B) come from the same experiments; the data were divided into groups that were cultured in the presence (A) and absence (B) of a 5-HT1A agonist, and data from basal/unstimulated control (C) and ethanol-exposed cultures (E) are repeated in both figures. In these studies there was a significant main effect of ethanol [F(1,102) = 47.1, P < 0.0001]. With the exception of the EW data point in (B) (mean of 2 experiments), all values are the mean T the SEM of values obtained from 3 to 11 separate experiments.

M. Druse et al. / Developmental Brain Research 159 (2005) 18 – 28

23

3.3. The protective effects of 5-HT1A agonists are blocked by LY294002, an inhibitor of PI-3K

Fig. 3. There is a modest reduction in pAkt levels in cultures that were treated with 50 mM ethanol for 24 h. (A) Includes a representative Western blot along with the corresponding loading control: Coomassie blue-stained actin bands. (B) is a graphic summary of data from 9 experiments. A comparison of relative O.D. data from the individual experiments demonstrated a significant effect of ethanol [F(1,17) = 8.1, P < 0.05, n = 9]. The effects of ethanol on pAkt are presented as the mean percentage of control values T the SEM. The ** indicate that the values from ethanol-treated animals were significantly reduced at P < 0.01.

control cultures treated with ipsapirone or 8-OH-DPAT were, respectively, 78.5% T 4.6% and 78.8% T 5.0%; values for ethanol-exposed cultures treated with the same drugs were, respectively, 70.6% T 8.6% and 72.4% T 4.3%. 3.2. The protective effects of 5-HT1A agonists on the survival of ethanol-treated fetal rhombencephalic neurons are antagonized by the 5-HT1A antagonist WAY 100635 (Fig. 2A) In the presence of WAY 100635, the percentage of surviving neurons in cultures that were co-treated with ethanol and a 5-HT1A agonist (E1AW) was significantly less than that found in the absence of WAY 100635 (E1A) (P < 0.01). The percentage of surviving/non-apoptotic neurons in control cultures was also reduced in the presence of WAY 100635 (Fig. 2A: CIA vs. CIAW, P < 0.01; Fig. 2B: C vs. CW, P < 0.01). However, WAY 100635 did not further reduce the percentage of surviving neurons in ethanoltreated cultures (Fig. 2B: E vs. EW, P > 0.05).

LY294002 reduced the percentage of surviving neurons, co-treated with both ethanol and a 5-HT1A agonist to a value that was comparable to that in ethanol-treated cultures (Fig. 2A: E1A vs. E1ALY, P < 0.01). In comparison with control cultures, treatment with LY294002 also significantly reduced the percentage of surviving/non-apoptotic neurons in cultures that were maintained both in the absence (Fig. 2B: C vs. CLY P < 0.01) and presence (Fig. 2A: C1A vs. C1ALY, P < 0.05) of a 5-HT1A agonist; these reduced levels were comparable to those in cultures of ethanol-treated neurons. LY294002 did not further reduce the percentage of surviving neurons in ethanol-treated cultures (Fig. 2B, P < 0.05). 3.4. The protective effects of 5-HT1A agonists were inhibited by PD98059, an inhibitor of MAPKK (Fig. 2A: E1A vs. E1APD) Although PD98059 also reduced the percentage of surviving neurons in control cultures (Fig. 2B: CPD vs. C, P < 0.05), this enzyme inhibitor did not exert a significant effect in control cultures that were treated with a 5-HT1A agonist (Fig. 2A: C1A vs. C1APD P > 0.05). In addition, PD98059 did not further reduce the percentage of surviving neurons in ethanoltreated cultures (Fig. 2B: E vs. EPD, P > 0.05). The graphic summary of the data presented in Fig. 3 demonstrates that there is a modest reduction in pAkt levels in cultures that were treated with 50 mM ethanol for 24 h. Analyses of nine separate experiments found that ethanol treatment reduced basal levels of pAkt by ¨30%. A comparison of relative optical density (O.D.) data from multiple individual experiments demonstrated a significant effect of ethanol [ F(1,17) = 8.1, P < 0.05], and the post hoc analysis showed that the relative O.D. for pAkt in ethanoltreated neurons was significantly reduced (P < 0.01). The effects of ethanol on the levels of pAkt are presented as a percentage of control values.

Fig. 4. PI-3KYpAkt is stimulated by a 10-min treatment with ipsapirone (Ips; A, B) stimulates PI-3KY pAkt in cultures of control (Con) and ethanol-treated neurons (Eth). (A) Includes a representative Western blot and the corresponding Coomassie blue-stained actin bands. (B) is a graphic depiction of the summary of densitometric data obtained from multiple studies with values presented as a percentage of control values. Means and SEM were obtained from Western blots of samples from 3 to 10 separate experiments.

24

M. Druse et al. / Developmental Brain Research 159 (2005) 18 – 28

3.5. 5-HT1A agonists activate PI-3K-mediated phosphorylation of Akt in fetal rhombencephalic neurons As shown in Fig. 4, a 10-min stimulation of both control and ethanol-treated fetal rhombencephalic neurons with the 5-HT1A agonist ipsapirone (Figs. 4A and 4B, n = 10) increased the phosphorylation of Akt. In additional experiments, we also found that treatment with 100 nM of buspirone or 8-OH-DPAT also stimulated the phosphorylation of Akt to 200% of control values (data not shown). In this and subsequent pAkt figures, we used a 10-min stimulation period because preliminary analyses of time courses indicated that ipsapirone-stimulated phosphorylation of Akt peaked at ¨5 to10 min. Levels of pAkt typically declined to those in unstimulated cultures by ¨20 to 30 min after stimulation (data not shown). As shown in Fig. 5, ipsapirone-induced phosphorylation of Akt is inhibited by LY294002. 3.6. Neither a 24-h treatment of fetal rhombencephalic neurons with 50 mM ethanol nor a 5- to 10-min stimulation of these neurons with a 5-HT1A agonist significantly altered the level of p42/p44 MAPK (Fig. 6) The relative densities of p42/p44 MAPK from 11 different ethanol-treated and cultures of fetal rhombencephalic neurons were compared to those of the paired control cultures. This comparison demonstrated that a 24-h treatment of ethanol did not exert a significant effect; ethanol values were 111 T 15% of control values. In order to determine whether a 5-HT1A agonist significantly increased p42/p44 MAPK, we stimulated these neuronal cultures with ipsapirone for 5 or 10 min: times that we determined to correspond to the highest signal for p42/p44MAPK. At 5 and 10 min, respectively, the relative densities of p42/p44 MAPK in ipsapirone-stimulated cultures that were maintained in control media were 111 T 4% (n = 7) and 102 T 15% (n = 7) of control values. Following a 5-min stimulation with

Fig. 6. Effects of ethanol and 5-HT1A agonists on p42//p44MAPK in fetal rhombencephalic neurons. (B) Provides a graphic depiction of the effects of a 24-h treatment with 50 mM ethanol on the relative density of p42/p44 MAPK, and (D) depicts the effects of a 5- or 10-min treatment with ipsapirone. (A, C) Include representative Western blots and the corresponding Coomassie blue-stained actin bands for (B, D), respectively. When we compared the results obtained from 11 different paired experiments, we found that the relative density of p42/p44 MAPK in ethanol-treated cultures was 111 T 15% of control values (B). In addition, we found that the relative density of p42/p44 MAPK in control cultures treated for 5 or 10 min with ipsapirone were, respectively, 111 T 4% (n = 7) and 102 T 15% (n = 7) of that in unstimulated control cultures (D). Following a 5-min stimulation with ipsapirone, the values in ethanol-treated cultures were 113 T 33%.

ipsapirone, the values in ethanol-treated cultures were 113 T 33% (data not shown). In addition, we found that ipsapironestimulated phosphorylation of p42/p44 MAPK for 1 h was 107 T 21% (n = 5) in cultures maintained in control media and 111 T 8% (n = 5) in cultures maintained in ethanolcontaining media. Similar results were found in cultures maintained in control and ethanol-treated media that were stimulated with 8-OH-DPAT for 1 h (122 T 28%, n = 4, and 89 T 17%, n = 3) (Fig. 6).

4. Discussion

Fig. 5. 5-HT1A stimulation of pAkt is inhibited by LY294002. (A) Depicts a representative Western blot and the corresponding Coomassie blue-stained actin bands. (B) Provides a graphic depiction of the relative O.D.’s of each band as presented as a percentage of control values. The following abbreviations were used: control (C), control + ipsapirone (CI), control + ipsapirone + LY294002 (CILY).

4.1. In comparison with control cultures, treatment with 50 mM ethanol induced apoptosis and thus reduced the percentage of surviving/non-apoptotic neurons from ¨78% to ¨56% These findings and observations from an earlier study [21] emphasize the vulnerability of fetal rhombencephalic neurons to ethanol [23,56,59,60,64]; they also suggest that

M. Druse et al. / Developmental Brain Research 159 (2005) 18 – 28

apoptosis is a cause of the ethanol-associated reduction in 5HT neurons [21,56]. Additional in vitro studies detect ethanol-induced apoptosis in cerebellar granule cells [48], cortical neurons [34], cortical explants [13], and hippocampal neurons [47]. In vivo studies show that ethanol promotes apoptosis in the developing neural crest [11,22,42]. Ethanol also causes neurodegeneration during late fetal and early postnatal brain development [33] and causes mitochondrial dysfunction, which can result in apoptotic cell death in fetal brain [51]. 4.2. The present study provides additional evidence of the potent anti-apoptotic effects of 5-HT1A agonists In a prior report, we demonstrated the anti-apoptotic effects of ipsapirone [21]; in this study, we confirmed our earlier findings and showed that the 5-HT1A agonist, e.g., 8OH-DPAT, also prevented ethanol-associated apoptosis in fetal rhombencephalic neurons. Both 8-OH-DPAT and another 5-HT1A agonist, i.e., Bay X 3702, attenuate apoptosis in neurons from additional brain areas (i.e., hippocampus and cortex) when apoptosis is induced by staurosporine treatment [5], serum deprivation [3,4], and glutamate excitotoxicity [5]. 8-OH-DPAT also suppresses anoxia-induced apoptosis in neuronal HN2-5 cells [1]. It is hypothesized that the anti-apoptotic/protective effects of 5-HT1A agonists on fetal 5-HT neurons prevent the ethanol-associated reduction of 5-HT [19,52,58], and that restoration of normal levels of 5-HT promotes the normal neurotrophic and developmental effects of 5-HT on serotonergic and other CNS neurons [39,63]. In vivo the trophic actions of 5-HT are likely to be mediated by somatodendritic 5-HT1A receptors on raphe nuclei [28,62] and by 5-HT1A receptors on astrocytes [8,61]. 4.3. The anti-apoptotic effects of the 5-HT1A agonists 8-OH-DPAT and ipsapirone appear to be specific for the 5-HT1A receptor, because they are blocked by the 5-HT1A receptor antagonist WAY 100635 Although WAY 100635 also caused a 19% reduction of ‘‘basal’’ survival, e.g., survival in control cultures that were not treated with a 5-HT1A agonist, it did not further reduce the percentage of fetal rhombencephalic neurons in ethanoltreated cultures. There are two conclusions that can be made from this observation. First, WAY 100635 did not exert toxic effects on ethanol cultures; its effects on ethanoltreated cultures appear to be limited to the antagonism of the actions of a 5-HT1A agonist. Secondly, the effects of WAY 100635 on control cultures may be explained by its antagonism of the actions of 5-HT that is secreted into the media by control neurons. Secreted 5-HT is expected to act in an autocrine manner on 5-HT1A receptors to promote survival of 5-HT neurons. In contrast, ethanol-treated cultures lack may sufficient secreted 5-HT to stimulate 5HT1A receptor-mediated pro-survival effects. The latter

25

hypothesis is supported by our earlier in vivo and in vitro studies. In these studies, this laboratory demonstrated that the concentration of 5-HT was reduced in the fetal brain area of ethanol-exposed offspring that contained the cell bodies of developing 5-HT neurons; 5-HT was reduced in the offspring of ethanol-exposed rats by nearly 50% at G15 and by 65% at G19 [19]. In addition, both in vivo and in vitro ethanol exposure reduced 5-HT neurons. There was a 40% reduction in 5-HT neurons in the dorsal raphe of postnatal day 5 ethanol-exposed offspring [59], and a 40% reduction in 5-HT neurons in fetal rhombencephalic tissue that was cultured under the same conditions as those used in the present study [21]. 4.4. It appears that the protective effects of a 5-HT1A agonist on primary cultures of fetal rhombencephalic neurons are mediated by the PI-3KYpAkt pathway, because an inhibitor of PI-3K, i.e., LY294002, blocked these effects of 5-HT1A agonists Although LY294002 did not reduce basal survival or exert toxic effects on ethanol-treated cultures, it did inhibit basal survival by 18% in control cultures, presumably by inhibiting PI-3K. Similar to our previous discussion of WAY 100635, the different effects of LY294002 on basal survival in control and ethanol-treated cultures might be explained by the differences in the concentration of secreted 5-HT in the media and consequently to differences in basal activation of 5-HT1A receptor-mediated pro-survival effects. The hypothesis that 5-HT1A agonist-mediated survival involves the activation of the PI-3K pathway is also supported by Western blot analyses, which showed that ipsapirone increases pAkt and that LY294002 blocks this effect. Moreover, ipsapirone activates PI-3KYpAkt in ethanol-treated as well as in control neurons. Although we are unaware of additional evidence in primary neurons, prior investigations using cell lines [1,16,26,53] and a study of the brains of suicide victims [32] link 5-HT1A agonists with the activation of the PI-3K pathway. Clearly, there are cellular differences in the signaling and survival pathways that are activated by 5-HT1A agonists. Another investigation reported that although PI-3Kg was activated by 8-OH-DPAT in neuronal HN2-5 cells, this activation was not linked with 5-HT1A agonist-mediated suppression of anoxia-induced apoptosis in HN2-5 cells [1]. Work is underway to determine which of the downstream effectors of this pathway are specifically involved with 5-HT1A-mediated survival of ethanol-treated fetal rhombencephalic neurons. 4.5. The present report suggests that the MAPKK pathway might be involved with the 5-HT1A agonist-mediated protective/anti-apoptotic effects in primary cultures of fetal rhombencephalic neurons This evidence comes from studies of Hoechst-stained neurons that were cultured with 5-HT1A agonists in the

26

M. Druse et al. / Developmental Brain Research 159 (2005) 18 – 28

presence of the MAPKK inhibitor PD98059 for 24 h. However, the Western blot analyses did not detect a marked increase of p42/p44 MAPK in fetal rhombencephalic neurons following stimulation with a 5-HT1A agonist for intervals ranging from 5 to 60 min. Prior studies in cell lines report a link between 5-HT1A receptors and MAPK [26,53]. In fact, in the neuronal HN2-5 cell line, there is evidence that the MAPK pathway accounts for anti-apoptotic effects of 8-OH-DPAT on anoxia-induced apoptosis and for the suppression of caspase-3 activity [1,2]. It is not clear why the present studies did not detect a significant increase in p42/p44 MAPK in fetal rhombencephalic neurons by a 5-HT1A agonist. Clearly, these neurons respond to 5-HT1A agonists with regard to survival (this study; [21]) and stimulation of pAkt (this study). In addition, we detect p42/p44 MAPK in these neurons. Moreover, PD90589 inhibited 5-HT1A agonist-mediated survival (this study) and reduced basal levels of p42/p44 MAPK in untreated control cultures (data not shown). It is possible there is a modest though non-significant stimulation by 5-HT1A agonists of the p42/p44 MAPK pathway in fetal rhombencephalic neurons. Although such a link might not be fully elaborated in fetal neurons, the link might be enhanced at a later time in development. Interestingly, there is evidence of a possible link between the 5-HT1A-mediated activation of PI-3K and MAPK; inhibitors of both PI-3K and phosphatidylcholine hydrolysis caused a partial reduction of MAPK in transfected cells [16]. 4.6. A 24-h treatment of fetal rhombencephalic neurons with 50 mM ethanol significantly reduced the levels of pAkt. It is possible that this decrease partly explains the reduced survival and increased apoptosis in these neurons. However, ethanol exerted only a modest and insignificant effect on MAPK in these neurons. A review of the literature indicates that ethanol treatment can increase [36,57], decrease [36,55], or have no effect [57] on p42/p44 MAPK in neural tissue. Moreover, a recent study [12] found that the effects of ethanol on the activation of activity-dependent pMAPK occur in an age- and brain region-specific manner. Thus, the apparently disparate effects of ethanol on MAPK reported in the literature can likely be explained by differences in the use of brain, specific brain regions, neurons, or astrocytes; animal ages; in vivo, in vitro, acute, or chronic ethanol treatments; or the selected doses of ethanol. 4.7. In summary, 50 mM ethanol (¨230 mg/dl) markedly reduces survival and increases apoptosis in the population of neurons, i.e., fetal rhombencephalic neurons, that contain the developing 5-HT neurons Ethanol-associated apoptosis can be suppressed by treatment with a 5-HT1A agonist. The signaling pathway(s) involved with the anti-apoptotic effects of 5-HT1A agonists

in fetal rhombencephalic neurons appears to include the PI3K pathway. Evidence is provided that the MAPKK pathway might also be involved.

Acknowledgments This research was supported by a grant from the USPHSAA03490, Loyola University Chicago Stritch School Endowment from the William G. Potts Estate, and an Illinois Excellence in Academic Medicine Grant.

References [1] T. Adayev, Y. El-Sherif, M. Barua, N.J. Penington, P. Banerjee, Agonist stimulation of the serotonin1A receptor causes suppression of anoxia-induced apoptosis via mitogen-activated protein kinase in neuronal HN2-5 cells, J. Neurochem. 72 (1999) 1489 – 1496. [2] T. Adayev, I. Ray, R. Sondhi, T. Sobocki, P. Banerjee, The G proteincoupled 5-HT1A receptor causes suppression of caspase-3 through MAPK and protein kinase Ca, Biochim. Biophys. Acta 1640 (2003) 85 – 96. [3] B. Ahlemeyer, J. Krieglstein, Stimulation of 5 HT1A receptor inhibits apoptosis induced by serum deprivation in cultured neurons from chick embryo, Brain Res. 777 (1997) 179 – 186. [4] B. Ahlemeyer, A. Glaser, C. Schaper, I. Semkova, J. Krieglstein, The 5-HT1A receptor agonist Bay X 3702 inhibits apoptosis induced by serum deprivation in cultured neurons, Eur. J. Pharmacol. 370 (1999) 211 – 216. [5] B. Ahlemeyer, H. Beier, I. Semkova, C. Schaper, J. Krieglstein, S100beta protects cultured neurons against glutamate and staurosporine-induced damage and is involved in the antiapoptotic action of the 5-HT1A receptor agonist, Bay X 3702, Brain Res. 858 (2000) 121 – 128. [6] R. Andrade, R. Malenka, R.A. Nicoll, A G protein couples serotonin and GABA-B receptors to the same channels in hippocampus, Science 234 (1986) 1261 – 1265. [7] D.J. Arndt-Jovin, T.M. Jovin, Analysis and sorting of living cells according to deoxyribonucleic acid content, J. Histochem. Cytochem. 25 (1977) 585 – 589. [8] E.C. Azmitia, P.J. Gannon, N.M. Kheck, P.M. Whitaker-Azmitia, Cellular localization of the 5-HT1A receptor in primate brain neurons and glial cells, Neuropsychopharmacology 14 (1996) 35 – 46. [9] D.J. Bonthius, N.E. Bonthius, R.M. Napper, S.J. Astley, S.K. Clarren, J.R. West, Purkinje cell deficits in nonhuman primates following weekly exposure to ethanol during gestation, Teratology 53 (1996) 230 – 236. [10] G.J. Brewer, J.R. Torricelli, E.K. Evege, P.J. Price, Optimized survival of hippocampal neurons in B27-supplemented neurobasal, a new serum-free medium combination, J. Neurosci. Res. 35 (1993) 567 – 576. [11] M.M. Cartwright, S.M. Smith, Increased cell death and reduced neural crest cell numbers in ethanol-exposed embryos: partial basis for the fetal alcohol syndrome phenotype, Alcohol.: Clin. Exp. Res. 19 (1995) 378 – 386. [12] J.L. Chandler, G. Sutton, Acute ethanol inhibits extracellular signalregulated kinase, protein kinase B, and adenosine 3V:5V-cyclic monophosphate response element binding protein activity in an age- and brain region-specific manner, Alcohol.: Clin. Exp. Res. 29 (2005) 672 – 682. [13] Z.F. Cheema, J.R. West, R.C. Miranda, Ethanol induces Fas/Apo [apoptosis]-1 mRNA and cell suicide in the developing cerebral cortex, Alcohol.: Clin. Exp. Res. 24 (2000) 535 – 543.

M. Druse et al. / Developmental Brain Research 159 (2005) 18 – 28 [14] S.K. Clarren, D.W. Smith, The fetal alcohol syndrome, N. Engl. J. Med. 298 (1978) 1063 – 1067. [15] C.D. Coles, K.A. Platzman, C.L. Raskind-Hood, R.T. Brown, A. Falek, I.E. Smith, A comparison of children affected by prenatal alcohol exposure and attention deficit, hyperactivity disorder, Alcohol.: Clin. Exp. Res. 21 (1997) 150 – 161. [16] D.S. Cowen, R.S. Sowers, D.R. Manning, Activation of a mitogenactivated protein kinase (ERK2) by the 5-hydroxytryptamine1A receptor is sensitive not only to inhibitors of phosphatidylinositol 3kinase, but to an inhibitor of phosphatidylcholine hydrolysis, J. Biol. Chem. 271 (1996) 22297 – 22300. [17] F. DeVitry, M. Hamon, J. Catelon, M. Bubois, J. Thibault, Serotonin initiates and autoamplifies its own synthesis during mouse central nervous system development, Proc. Natl. Acad. Sci. 83 (1986) 8629 – 8633. [18] M. De Vivo, S. Maayani, Characterization of the 5-hydroxytryptamine-1A receptor-mediated inhibition of forskolin-stimulated adenylate cyclase activity in guinea pig and rat hippocampal membranes, J. Pharmacol. Exp. Ther. 238 (1987) 248 – 253. [19] M.J. Druse, A. Kuo, N. Tajuddin, Effects of in utero ethanol exposure on the developing serotonergic system, Alcohol.: Clin. Exp. Res. 15 (1991) 678 – 684. [20] M.J. Druse, H.-L. Hao, J.L. Eriksen, In utero ethanol exposure increases proenkephalin, a precursor of a neuropeptide that is inhibitory to neuronal growth, Alcohol.: Clin. Exp. Res. 23 (1999) 1519 – 1527. [21] M.J. Druse, N.F. Tajuddin, R.A. Gillespie, E. Dickson, M. Atieh, C.A. Pietrzak, P.T. Le, The serotonin-1A agonist ipsapirone prevents ethanol-associated death of total rhombencephalic neurons and prevents the reduction of fetal serotonin neurons, Brain Res. Dev. Brain Res. 15 (2004) 79 – 88. [22] W.C. Dunty, S.Y. Chen, R.M. Zucker, D.B. Dehart, K.K. Sulik, Selective vulnerability of embryonic cell populations to ethanolinduced apoptosis: implications for alcohol-related birth defects and neurodevelopmental disorders, Alcohol.: Clin. Exp. Res. 25 (2001) 1523 – 1535. [23] J.L. Eriksen, M.J. Druse, Astrocyte-mediated trophic support of developing serotonin neurons: effects of ethanol, buspirone, and S100B, Brain Res. Dev. Brain Res. 131 (2001) 9 – 15. [24] J.L. Eriksen, R.A. Gillespie, M.J. Druse, Effects of ethanol and 5HT1A agonists on astroglial S100B, Brain Res. Dev. Brain Res. 139 (2002) 97 – 105. [25] C. Famy, A.P. Streissguth, A.S. Unis, Mental illness in adults with fetal alcohol syndrome or fetal alcohol effects, Am. J. Psychiatry 155 (1998) 552 – 554. [26] M.N. Garnavskoya, T. van Biesen, B. Hawe, S. Casanos Ramos, J.R. Lefkowitz, J.R. Raymond, Ras-dependent activation of fibroblast mitogen-activated protein kinase by 5-HT1A receptor via a G protein beta gamma-subunit-initiated pathway, Biochemistry 35 (1996) 13716 – 13722. [27] C.R. Goodlett, A.T. Eilers, Alcohol-induced Purkinje cell loss with a single binge exposure in neonatal rats: a stereological study of temporal windows of vulnerability, Alcohol.: Clin. Exp. Res. 21 (1997) 738 – 744. [28] M.D. Hall, S. El Mestikawy, M.B. Emerit, L. Pichat, M. Hamon, H. Gozlan, [3H]8-hydroxy-2-(di-N-propylamino)tetralin binding to preand postsynaptic 5-hydroxytryptamine sites in various regions of the brain, J. Neurochem. 44 (1985) 1685 – 1696. [29] P.G. Haydon, D.P. McCobb, S.B. Kater, The regulation of neurite outgrowth, growth cone motility, and electrical synaptogenesis by serotonin, J. Neurobiol. 18 (1987) 197 – 215. [30] P. Honegger, F. Monnet-Tschudi, Aggregating neural cell cultures, in: S. Federoff, A. Richardson (Eds.), Protocols for Neural Cell Culture, 2nd ed., Humana Press, Totowa, NJ, 1997, pp. 25 – 49. [31] D. Hoyer, D.E. Clarke, J.R. Fozard, P.R. Hartig, G.R. Martin, E.J. Myelcharane, P.R. Saxena, P.P. Humphrey, International Union of

[32]

[33]

[34]

[35]

[36]

[37]

[38]

[39]

[40] [41] [42] [43]

[44]

[45]

[46]

[47]

[48]

[49]

[50]

[51]

27

Pharmacology classification of receptors of 5-hydroxytryptamine (serotonin), Pharmacol. Rev. (1994) 157 – 203. S.C. Hsiung, M. Adlersberg, V. Arango, J.J. Mann, H. Tamir, K.P. Liu, Attenuated 5-HT1A receptor signaling in brains of suicide victims: involvement of adenylyl cyclase, phosphatidylinositol 3kinase, Akt, and mitogen-activated protein kinase, J. Neurochem. 87 (2003) 182 – 194. C. Ikonomidou, P. Brittigau, M.J. Ishimaru, D.F. Wozniak, C. Koch, K. Genz, M.T. Price, V. Stefovska, F. Horster, T. Tenkova, K. Dikranian, J.W. Olney, Ethanol-induced apoptotic neurodegeneration and fetal alcohol syndrome, Science 287 (2000) 1056 – 1060. J.S. Jacobs, M.W. Miller, Proliferation and death of cultured fetal neocortical neurons: effects of ethanol on the dynamics of cell growth, J. Neurocytol. 30 (2001) 391 – 401. S.W. Jacobson, J.L. Jacobson, R.J. Sokol, S.S. Martier, J.W. Ager, Prenatal alcohol exposure and infant information processing ability, Child. Dev. 64 (1993) 1706 – 1721. H.S. Kalluri, M.K. Ticku, Regulation of ERK phosphorylation by ethanol in fetal cortical neurons, Neurochem. Res. 28 (2003) 765 – 769. J.-A. Kim, M.J. Druse, Protective effects of maternal buspirone treatment on serotonin reuptake sites in ethanol-exposed offspring, Brain Res. Dev. Brain Res. 92 (1996) 190 – 198. N. Konig, M.B. Wilkie, J. Lauder, Dissection of monoaminergic neuronal groups from embryonic rat brain, in: A. Sharar, J. de Vellis, A. Vernadakis, B. Haber (Eds.), A Dissection and Tissue Culture Manual of the Nervous System, Alan R. Liss, Inc., New York, 1989, pp. 26 – 29. J.M. Lauder, Ontogeny of the serotonergic system in the rat: serotonin as a developmental signal, Ann. N. Y. Acad. Sci. 600 (1990) 297 – 314. J.M. Lauder, H. Krebs, Serotonin as a differentiation signal in early neurogenesis, Dev. Neurosci. 1 (1978) 15 – 30. J.M. Lauder, J.A. Wallace, M.B. Wilkie, A. DiNome, H. Krebs, Roles for serotonin in neurogenesis, Monogr. Neural Sci. 9 (1983) 3 – 10. P. Liesi, Ethanol-exposed central neurons fail to migrate and undergo apoptosis, J. Neurosci. Res. 48 (1997) 439 – 448. K.E. Light, S.M. Belcher, D.R. Pierce, Time course and manner of Purkinje neuron death following a single ethanol exposure on postnatal day 4 in the developing rat, Neuroscience 114 (2002) 327 – 337. S.E. Maier, J.A. Miller, J.M. Blackwell, J. West, Fetal alcohol exposure and temporal vulnerability: regional differences in cell loss as a function of the timing of binge-like alcohol exposure during brain development, Alcohol.: Clin. Exp. Res. 23 (1999) 726 – 734. S.N. Mattson, A.M. Schoenbeld, E.P. Riley, Teratogenic effects of alcohol on brain and behavior, Alcohol Res. Health 25 (2001) 185 – 191. M.W. Miller, Generation of neurons in the rat dentate gyrus and hippocampus: effects of prenatal and postnatal treatment with ethanol, Alcohol.: Clin. Exp. Res. 19 (1995) 1500 – 1509. K.L. Moulder, T. Fu, H. Melbostad, R.J. Cormier, K.E. Isenberg, C.F. Zorumski, S. Mennerick, Ethanol-induced death of postnatal hippocampal neurons, Neurobiol. Dis. 10 (2002) 396 – 409. J. Oberdoerster, R.A. Rabin, Enhanced caspase activity during ethanol-induced apoptosis in rat cerebellar granule cells, Eur. J. Pharmacol. 385 (1999) 273 – 282. N.J. Pantazis, J.R. West, D. Dai, The nitric oxide-cyclic GMP pathway plays an essential role in both promoting cell survival of cerebellar granule cells in culture and protecting the cells against ethanol neurotoxicity, J. Neurochem. 70 (1998) 1826 – 1838. N.J. Penington, J.S. Kelly, A.P. Fox, A study of the mechanism of Ca2+ current inhibition produced by serotonin in rat dorsal raphe neurons, J. Neurosci. 11 (1991) 3594 – 3609. V. Ramachandran, A. Perez, J. Chen, D. Senthil, S. Schenker, G.I. Henderson, In utero ethanol exposure causes mitochondrial dysfunction, which can result in apoptotic cell death in fetal brain: a

28

[52]

[53]

[54]

[55]

[56]

[57]

[58]

M. Druse et al. / Developmental Brain Research 159 (2005) 18 – 28 potential role for 4-hydroxynonenal, Alcohol.: Clin. Exp. Res. 25 (2001) 862 – 871. W. Rathbun, M.J. Druse, Dopamine, serotonin and acid metabolites in brain regions from the developing offspring of ethanol treated rats, J. Neurochem. 44 (1985) 57 – 62. J.R. Raymond, Y.V. Mukhim, T.W. Gettys, M.N. Garnovskaya, The recombinant 5-HT1A receptor: G protein coupling and signaling pathways, Br. J. Pharmacol. 127 (1999) 1751 – 1764. T.M. Roebuck, S.N. Mattson, E.P. Riley, Behavioral and psychosocial profiles of alcohol-exposed children, Alcohol.: Clin. Exp. Res. 23 (1999) 1070 – 1076. P.P. Sanna, C. Simpson, R. Lutjens, G. Koob, ERK regulation in chronic ethanol exposure and withdrawal, Brain Res. 948 (2002) 186 – 191. Y. Sari, F.C. Zhou, Prenatal alcohol exposure causes long-term serotonin neuron deficit in mice, Alcohol.: Clin. Exp. Res. 28 (2004) 941 – 948. T.L. Smith, E. Navratilova, The effect of ethanol exposure on mitogenactivated protein kinase activity and expression in cultured rat astrocytes, Neurosci. Lett. 341 (2003) 91 – 94. N.F. Tajuddin, M.J. Druse, Treatment of pregnant alcohol-consuming rats with buspirone: effects on serotonin and 5-hydroxyindoleacetic acid content in offspring, Alcohol.: Clin. Exp. Res. 17 (1993) 110 – 114.

[59] N.F. Tajuddin, M.J. Druse, In utero ethanol exposure decreased the density of serotonin neurons. Maternal ipsapirone treatment exerted a protective effect, Brain Res. Dev. Brain Res. 117 (1999) 91 – 97. [60] N.F. Tajuddin, M.J. Druse, A persistent deficit of serotonin neurons in the offspring of ethanol-fed dams: protective effects of maternal ipsapirone treatment, Brain Res. Dev. Brain Res. 129 (2001) 181 – 188. [61] P.M. Whitaker-Azmitia, J.M. Lauder, A. Shenner, E.C. Azmitia, Postnatal changes in serotonin1 receptors following prenatal alterations in serotonin levels: further evidence for functional fetal serotonin1 receptors, Dev. Brain Res. 33 (1987) 285 – 289. [62] P.M. Whitaker-Azmitia, R. Murphy, E.C. Azmitia, Stimulation of astroglial 5-HT1A receptors releases the serotonergic growth factor, protein S-100, and alters astroglial morphology, Brain Res. 528 (1990) 155 – 158. [63] P.M. Whitaker-Azmitia, M. Druse, P. Walker, J.M. Lauder, Serotonin as a developmental signal, Behav. Brain Res. 73 (1996) 19 – 29. [64] F.C. Zhou, Y. Sari, J.K. Zhang, C.R. Goodlett, T. Li, Prenatal alcohol exposure retards the migration and development of serotonin neurons in fetal C57BL mice, Dev. Brain Res. 125 (2001) 147 – 155. [65] F.C. Zhou, Y. Sari, T.K. Li, C. Goodlett, E.C. Azmitia, Deviations in brain early serotonergic development as a result of fetal alcohol exposure, Neurotox. Res. 4 (2002) 337 – 342.