Estrogen-mediated regulation of cholinergic expression in basal forebrain neurons requires extracellular-signal-regulated kinase activity

Neuroscience 124 (2004) 809 – 816

ESTROGEN-MEDIATED REGULATION OF CHOLINERGIC EXPRESSION IN BASAL FOREBRAIN NEURONS REQUIRES EXTRACELLULARSIGNAL-REGULATED KINASE ACTIVITY J. L. PONGRAC,a,b,c* D. B. DEFRANCOa

R.

B.

GIBBSb

AND

ner distinct from nerve growth factor and independent of improved survival. © 2004 IBRO. Published by Elsevier Ltd. All rights reserved.

a Department of Pharmacology, School of Medicine, School of Pharmacy, University of Pittsburgh, Pittsburgh, PA 15261, USA

Key words: estradiol, choline transport, ChAT, NGF, ACh, rat.

b

Pharmaceutical Sciences, School of Pharmacy, University of Pittsburgh, Pittsburgh, PA 15261, USA

Various lines of evidence suggest that estrogen may be a candidate for maintaining the cholinergic phenotype and sustaining the differentiated function of basal forebrain cholinergic neurons (BFCNs). In the rat, inputs from BFCNs contribute to estrogen-mediated changes in the hippocampus (Leranth et al., 2000; Rudick et al., 2003) and improvement in working memory (Daniel and Dohanich, 2001; Gibbs, 2002). BFCNs respond to estrogen treatment through improvement in markers of cholinergic function (Luine et al., 1985; O’Malley et al., 1987; Gibbs and Pfaff, 1992; Gibbs et al., 1994a, 1997; Gabor et al., 2003; Singh et al., 1994; Rabbini et al., 1997; Gibbs, 1997, 2000). In clinical studies, cortical but not striatal cholinergic terminal concentrations correlate with length of estrogen replacement in postmenopausal women (Smith et al., 2001). It is, however, still not known how estrogen mediates changes in the cholinergic phenotype. Evidence for estrogen acting directly on BFCNs is suggested by binding studies employing [3H]-estradiol (Toran-Allerand et al., 1992), the presence of nuclear estrogen receptor (ER) isoform ␣ (ER-␣) in nerve growth factor (NGF)-responsive BFCNs (Shughrue et al., 2000; Miettinen et al., 2002) and most recently the double-labeling of ER-␣ and vesicular acetylcholine transporters in axon terminals (Towart et al., 2003). The variety of ER sites predicts multiple mechanisms for the actions of estrogen on BFCN function including classical pathways directly mediating genomic changes or nonclassical, non-genomic pathways initiated through second messengers (Watters et al., 1997; Beyer and Raab, 1998; McEwen, 1999, 2001a,b; Vasudevan et al., 2001; Ivanova et al., 2001). The term non-genomic is used to distinguish between the mechanisms initiating the signal, but these pathways can also alter gene expression, albeit indirect, through kinase cascades following second messenger activation. The present study identifies one mechanism for estrogen-mediated regulation of cholinergic function in rat basal forebrain neurons distinct form NGF action and altered survival.

c

Department of Psychiatry, University of British Columbia, Vancouver General Hospital Research Pavilion, Room 311, 828 West 10th Avenue, Vancouver, BC, Canada V5Z 1L8

Abstract—Beyond the role estrogen plays in neuroendocrine feedback regulation involving hypothalamic neurons, other roles for estrogen in maintaining the function of CNS neurons remains poorly understood. Primary cultures of embryonic rat neurons together with radiometric assays were used to demonstrate how estrogen alters the cholinergic phenotype in basal forebrain by differentially regulating sodium-coupled high-affinity choline uptake and choline acetyltransferase activity. Highaffinity choline uptake was significantly increased 37% in basal forebrain cholinergic neurons grown in the presence of a physiological dose of estrogen (5 nM) from 4 to 10 days in vitro whereas choline acetyltransferase activity was not significantly changed in the presence of 5 or 50 nM estrogen from 4 to 10 or 10 to 16 days in vitro. Newly-synthesized acetylcholine was significantly increased 35% following 6 days of estrogen treatment (10 days in vitro). These effects are in direct contrast to those found for nerve growth factor; that is, nerve growth factor can enhance the cholinergic phenotype through changes in choline acetyltransferase activity alone [J Neurochem 66 (1996) 804]. This is most surprising given that mitogen-activated protein kinase and extracellular-signal-regulated kinase1/2, kinases also activated in the signaling pathway of nerve growth factor, were found to participate in the estrogen-mediated changes in the cholinergic phenotype. Likewise, general improvement in the viability of the cultures treated with estrogen does not account for the effects of estrogen as determined by lactate dehydrogenase release and nerve growth factor-responsiveness. These findings provide evidence that estrogen enhances the differentiated phenotype in basal forebrain cholinergic neurons through second messenger signaling in a man*Correspondence to: J. L. Pongrac, Department of Psychiatry, Vancouver General Hospital Research Pavilion, Room 311, 828 West 10th Avenue, Vancouver, BC, Canada V5Z 1L8. Tel: ⫹1-604-875-5259; fax: ⫹1-604-875-4376. E-mail address: [email protected] (J. L. Pongrac). Abbreviations: ACh, acetylcholine; BFCN, basal forebrain cholinergic neuron; BSA, bovine serum albumin; ChAT, choline acetyltransferase; DIV, days in vitro; E2, 17␤-estradiol; EDTA, ethylenediaminetetraacetic acid; ER, estrogen receptor; ERK, extracellular-signal-regulated kinase; HACU, high-affinity choline uptake; HC-3, hemicholinium-3; H2O2, hydrogen peroxide; HRP, horseradish peroxidase; LACU, lowaffinity choline uptake; LDH, lactate dehydrogenase; MEK, mitogenactivated protein kinase; NaCl, sodium chloride; NaOH, sodium hydroxide; NGF, nerve growth factor; PBS, phosphate-buffered saline; TCA, trichloroacetic acid.

EXPERIMENTAL PROCEDURES All tissue culture and reagent grade chemicals were from Sigma, St. Louis, MO, USA unless otherwise stated.

0306-4522/04$30.00⫹0.00 © 2004 IBRO. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.neuroscience.2004.01.013

809

810

J. L. Pongrac et al. / Neuroscience 124 (2004) 809 – 816

Culture of rat basal forebrain neurons All culture preparations followed the guidelines set by the Animal Welfare Act and the National Institute of Health Guide for the Care and Use of Laboratory Animals as approved by the Institutional Animal Care and Use Committee of the University of Pittsburgh. We made every effort to minimize the number of animals used and to avoid any suffering. Neuronal cultures from rat basal forebrain regions (septum, diagonal band of Broca and substantia innominata) were prepared from embryonic day 17 Sprague–Dawley rats removed from timed, pregnant mothers (Hilltop Laboratories, Scottdale, PA, USA) as described previously by Pongrac and Rylett (1996, 1998). Cultures were enriched for neurons by maintenance in serum- and phenol red-free Neurobasal medium: DMEM/F12 (3:2; Gibco Life Technologies, Grand Island, NY, USA) containing minimal supplementation (N2 supplements; Bottenstein and Sato, 1976; and 6 g/l L-glucose) and antibiotic (1U/ ml:1 ␮g/ml penicillin:streptomycin) and incubated at 37 °C in 5% CO2. One-third of the medium was replaced every 4 days. Treatments were replenished relative to the amount of medium replaced at feeding. Working solutions of 17␤-estradiol (E2; Sigma) were prepared as 5–50 ␮M E2 dissolved in ethanol vehicle and administered at 1000-fold concentration. Cultures not receiving estrogen were treated with vehicle alone at a final concentration of 0.0001% ethanol. Solutions of NGF (Harlan for BioScience Products, Madison, WI, USA) were prepared in phosphate-buffered saline (PBS; Life Technologies) containing 1% bovine serum albumin (BSA; PBS/BSA) and the vehicle control was PBS/BSA. Studies conducted with the reversible mitogen-activated protein kinase (MEK)1/2 inhibitor, U0126 (Cell Signaling Technology, Beverly, MA, USA), were treated with E2 for 5 min prior to receiving the inhibitor or dimethyl sulfoxide vehicle for 25 min. U0126 was removed by aspirating the medium and washing the cultures twice with fresh medium. Cultures were maintained in medium containing 5 nM E2 until they were assayed for high-affinity choline uptake (HACU) at 8 days in vitro (DIV). Note that basal forebrain neurons in vitro are prone to aggregate with advancing age of the culture and this is exacerbated with increased handling due to the washout procedure in this experiment. The earlier assay point was chosen to reduce complications arising from these procedures. Cultures treated with ICI 182,780 (Tocris, Ballwin, MO, USA) received E2 concurrently with a vehicle control of 0.0002% ethanol.

HACU assay HACU was assayed by measuring the amount of [3H]choline chloride taken up by cells over 5 min at 37 °C in the absence and presence of hemicholinium-3 (HC-3; Sigma) as described in Pongrac and Rylett (1996). HC-3-sensitive [3H]choline uptake represents neuron-specific high-affinity choline transport considered to be rate-limiting to the synthesis of acetylcholine (ACh; Jope, 1979). Neuronal cultures received 1.0 ␮M [methyl-3H]choline chloride (0.25 Ci/mmol; Dupont NEN Research Products, Boston, MA, USA) for 5 min in the presence and absence of 10 ␮M HC-3 and subsequently digested with 0.1 M sodium hydroxide (NaOH). An aliquot of the digest was dissolved in scintillation fluid and the tritium content was quantified by liquid scintillation spectrometry (counting efficiency, 54%). Sample protein was determined using Bio-Rad Protein Assay (Bio-Rad, Hercules, CA, USA) from aliquots of the digest.

Choline acetyltransferase (ChAT) assay Neurons were washed with PBS (pH 7.4) and frozen at ⫺80 °C until assayed. ChAT activity was measured using a modification of the radioenzymatic method of Fonnum (1969) as described by Pongrac and Rylett (1996). Cultures were thawed on ice and

incubated for 30 min in buffered phosphate containing EDTA (0.87 mM), eserine sulfate (0.15 mM) and Triton X-100 (0.5%) to lyse them. Each well was scraped and 7 ␮l of lysate were transferred in duplicate to fresh microfuge tubes. The 50 min reaction was initiated with 7 ␮l of 0.1 mM sodium phosphate buffer, pH 7.4 containing sodium chloride (NaCl) (0.15 mM), BSA (0.05%), eserine sulfate (0.1 mM), [3H]acetyl-CoA (0.43 Ci/mmol; ICN Radiochemicals, Irvine, CA, USA) and choline iodide (2 mM). A separate set of duplicate samples received the same incubation solution without choline iodide to correct for non-specific background. [3H]ACh was extracted and tritium quantified (counting efficiency, 50%). ChAT activity was expressed as nmol ACh synthesized/mg protein/h.

ACh synthesis Newly synthesized ACh was recovered from choline metabolites according to a modified method of Goldberg and McCaman (1973) as described by Pongrac and Rylett (1996). Choline uptake assay was followed to the point of digesting the cells, at which point the cells were digested with 1 ml of 5% trichloroacetic acid (TCA) instead of with NaOH. Plates were scraped and the contents transferred to a microcentrifuge. An aliquot of the acidic supernatant was used to determine the total [3H]choline uptake for each well and the remainder was extracted to recover [3H]ACh. An aliquot of the supernatant was prepared for extraction of [3H]choline metabolites by removing TCA with four volumes of watersaturated diethyl ether (four times). [3H]choline and choline esters including [3H]ACh were then extracted into tetraphenylboron salt in 3-heptanone followed by back-extraction of choline and ACh by shaking vigorously in an equal volume of 0.4 M HCl. The aqueous phase was reduced to dryness and the [3H]choline and [3H]ACh were separated by a modified method of Goldberg and McCaman (1973) as described by Pongrac and Rylett (1996). In brief, the samples were incubated with choline kinase (0.2 U/ml) at 37 °C for 30 min in a reaction mixture of 93 mM glycylglycine, pH 8.3, 15.5 mM ATP, 12 mM dithiothreitol and 18.5 mM MgCl2. The reaction was stopped and extracted with sodium tetraphenylboron in 3-heptanone (10 mg/ml) to extract [3H]ACh. Aliquots of the organic phase were added to liquid scintillant to determine the ACh by measuring tritium. [3H]ACh was expressed as d.p.m./5 min (counting efficiency, 50%).

Western blot analysis of phosphorylated extracellular-signal-regulated kinase (ERK) 1/2 Cultures treated with E2 were washed with ice-cold PBS at variable times and harvested for detection of ERK 1/2. Cells were collected via centrifugation (3500⫻g) and lysed on ice in 50 mM Tris, pH 7.5, 2 mM EDTA, 100 mM NaCl, 1% Nonidet-P40, 1 mM sodium orthovanadate, 1 ␮g/ml leupeptin and 2 ␮M dithiothreitol. Lysates were centrifuged at 12,000⫻g for 5 min at 4 °C and soluble extracts were collected for protein determinations. Western blot analyses were performed according to the method of Laemmeli (1970) by loading 25–30 ␮g into each well of a 10% sodium dodecyl sulfate–polyacrylamide gel. The proteins were blotted on PVDF membrane (Immobilon-P; Millipore, Bedford, MA, USA). Blots were blocked with 10% dried milk with 0.5% Tween 20 in PBS buffer for 1 h, incubated with anti-phosphoERK 1/2 antibody (1:500; Cell Signaling) overnight at 4 °C and incubated in horseradish peroxidase (HRP)-conjugated secondary anti-mouse antibody for 40 min at room temperature. PhosphoERK 1/2 was detected by chemiluminescence reaction (Western lightning kit; Perkin Elmer, Boston, MA, USA) and exposure to audioradiograph film. The blots were then stripped and reprobed with anti-ERK 1/2 (1:1500; Cell Signaling) and HRP-conjugated secondary anti-rabbit antibody following the same protocol as described above for anti-phosphoERK 1/2.

J. L. Pongrac et al. / Neuroscience 124 (2004) 809 – 816

HACU

A.

B.

Data analyses performed in duplicate or triplicate were averaged to give a single n value. Data as separate n values from multiple cultures (approximately 45 embryos/culture) were expressed as mean⫾S.E.M. and analyzed by paired t test, one-way repeated measures or two-way ANOVA followed by Neuman-Keuls multiple comparison test where appropriate (GraphPad Prism software, San Diego, CA, USA). The criterion for statistical significance was set at P⬍0.05.

Estrogen increases HACU and ACh synthesis without changing ChAT activity Primary cultures of rat basal forebrain neurons exposed to 5 nM estrogen from 4 to 10 DIV showed a 37% statistically significant increase in HACU as measured on 10 DIV (paired t test, two-tailed, P value⫽0.0153, n⫽8; Fig. 1A). Notably, the change in HACU was accompanied by a 35% change in newly-synthesized ACh (paired t test, two-tailed, P value⫽0.0362, n⫽5; Fig. 1B). Low-affinity choline uptake (LACU) as measured by [3H]choline taken up in the presence of HC-3 was not altered (nmoles [3H]choline/5 min/mg protein, mean⫾S.E.M.; control, 7.879⫾1.57; estradiol, 7.593⫾1.52; 93.6% control; paired t test, twotailed, P⫽0.4663, n⫽8). No changes in ChAT activity were measured from cultures treated with estrogen at 5 and 50 nM over the same culture period (E2 administered 4 –10 DIV, two-way ANOVA, P value⫽0.5067, n⫽5; Fig. 2A) or latter (E2 administered 10 –16 DIV, two-way ANOVA, P value⫽0.6520, n⫽5; Fig. 2B). ChAT activity, however, increased three- and seven-fold as a result of NGF treatment in cultures treated concurrently with NGF at 50 ng/ml and estrogen at 5 or 50 nM (two-way ANOVA, P⬍0.0001; Fig. 2A and B) demonstrating a pattern of continued development of NGF-responsive BFCNs in vitro. The changes in ChAT activity observed at 10 DIV are consistent with those previously reported for NGF at 8 DIV (Pongrac and Rylett, 1996). To determine whether estrogen-mediated changes in HACU require activation of ERs, ICI 182,780 was administered immediately prior to administering estrogen. ICI 182,780 is a pure antagonist of both isoforms of the ER that blocks ER activity in cortical neurons when administered at 100 times the dose of estrogen (Singh et al., 1999). Changes in HACU resulting in ⬎two-fold increase over control in the presence of estrogen were reduced to

a

3

0

0

5

(nmol)

Data analyses

RESULTS

6

17β -estradiol

ACh Synthesized

Neuronal cultures were pretreated with 5 nM E2 at 4 DIV. Hydrogen peroxide (H2O2) was serially diluted from 30% stock solution to a 1000-fold working solution with water. Cultures were exposed to H2O2 for 30 min at 8 DIV at which time the treatment was stopped by aspirating the H2O2-containing medium, the conditioned medium was replaced and cultures were returned to the incubator. Lactate dehydrogenase (LDH) release was assessed at 24 h after exposure to H2O2 using a Cytotoxicity Detection Kit (Boehringer Mannheim Biochemicals/Roche, Indiana, IL, USA) and a spectrophotometer at 490 nm. Cytotoxicity was determined according to the formula: cytotoxicity (%)⫽[(experimental value⫺untreated cell value)/(maximum LDH released from cells killed with Triton-X 100)]⫻100.

(nmol/5 min/mg protein)

Lactate dehydrogenase cytotoxicity analysis

811

1500

(dpm/5 min)

a 1000

500

0

0

5

17β -estradiol (nmol)

Fig. 1. Effect of estrogen on HACU and newly synthesized ACh. Rat basal forebrain neurons were treated with 5 nM estradiol at 4 DIV until assayed on 10 DIV for HACU expressed as nmoles [3H]choline taken up per mg of protein over 5 min and newly synthesized ACh expressed as d.p.m. [3H]ACh measured over 5 min. This resulted in (A) an 37% increase in HACU (mean⫾S.E.M., control 3.049⫾0.72, estradiol 4.140⫾0.81, n⫽8) and (B) an 35% increase in newly synthesized ACh (mean⫾S.E.M.; control, 687.4⫾114.6; estradiol 930.6⫾181.4, n⫽5) above control values. a Significantly different from vehicle control (P⬍0.05; paired t-test, two-tailed).

113% control when estrogen treatments were administered in the presence of 0.5 ␮M ICI 182,780 over two trials (nmoles [3H]choline/5 min/mg protein, mean⫾S.E.M.; control 3.061⫾1.35, E2 8.041⫾3.67, estradiol⫹ICI, 3.473⫾1.21). Estrogen-mediated changes in cholinergic markers regulated by ERK 1/2 kinase activity The small apparent decrease in ChAT activity observed when NGF and estrogen were administered together was not statistically significant when tested for interactions (two-way ANOVA; 10 DIV, P⫽0.5067, n⫽5; 16 DIV, P⫽0.6520, n⫽5) further suggesting that the mechanism of action for estrogen and NGF is distinct. Despite the absence of an interaction with NGF, basal forebrain neurons in culture showed an increase in the level of phosphoERK 1/2 relative to total ERK in response to 5 nM estrogen suggestive of an increase in ERK 1/2 activity (Fig. 3A) commonly observed in growth factor signaling. A selective MEK 1/2 inhibitor that reversibly blocks the phosphorylation of ERK 1/2, U0126, was administered at 10 ␮M to the basal forebrain neurons concurrent with estrogen at 4 DIV from 5 to 30 min following administration of estrogen.

812

J. L. Pongrac et al. / Neuroscience 124 (2004) 809 – 816

icant, dose-dependent increase in cytotoxicity that killed approximately 80% of the culture (two-way ANOVA, P⬍0.0001, n⫽4; Fig. 4). Estrogen at 5 nM, the concentration of estrogen required to increase HACU at 10 DIV, was not effective at protecting basal forebrain neurons (twoway ANOVA, P⫽0.5344, n⫽4; Fig. 4). Also, note there were no interactions between H2O2-mediated increase in LDH release and effects of 5 nM estrogen on these cultures (two-way ANOVA; E2 versus H2O2, P⫽0.6976).

ChAT Activity (nmol/min/mg protein)

A. 75

- NGF +NGF

50

25

0

0

5

50

0

5

50

17 β -Estradiol (nM) ChAT Activity (nmol/min/mg protein)

B. 75

50

25

0

0

5

50

0

5

50

17 β -Estradiol (nM) Fig. 2. Effect of estrogen and NGF on ChAT activity. Rat basal forebrain neurons were treated with 0, 5 or 50 nM estradiol at (A) four or (B) 10 DIV in the absence or presence of 50 ng/ml NGF with ChAT activity measured at 10 and 16 DIV, respectively. ChAT activity did not change at any dose of estrogen administered, whereas NGF-induced three- and seven-fold increases in ChAT activity above control values (10 DIV, P⬍0.0001, n⫽5; 16 DIV, P⬍0.0001, n⫽5; two-way ANOVA) demonstrating an enhanced NGF-responsiveness over time in culture.

U0126 blocked the estrogen-mediated increase in HACU at 8 DIV (one-way repeated measures ANOVA, P⫽0.0248, n⫽5; post hoc comparisons by Neuman-Keuls multiple comparison test; E2 vs control, P⬍0.05; E2 vs E2⫹U0126, P⬍0.05; control vs E2⫹U0126, P⬎0.05; Fig. 3B) presumably by blocking the phosphorylation of ERK 1/2. No changes in LACU were observed in response to estrogen (nmoles [3H]choline/5 min/mg protein, mean⫾S.E.M.; oneway repeated measures ANOVA, P⬎0.1552, n⫽5; control, 5.645⫾0.62; E2, 4.908⫾0.56; E2⫹U0126, 5.321⫾0.72). Estrogen-mediated changes in ACh were also attenuated by U0126 (one-way repeated measures ANOVA, P value⫽0.0483, n⫽6; post hoc comparisons by Neuman-Keuls multiple comparison test between control vs E2, P⬍0.05, estrogen vs E2⫹U0126, P⬍0.05, control vs E2⫹U0126, P⬎0.05; Fig. 3C). Estrogen does not increase survival of basal forebrain neurons under basal or oxidant stress conditions Culture viability was measured by the release and detection of LDH in the culture medium from 4 to 8 DIV in a two-way design where estrogen was administered in the presence and absence of an oxidant-stress in the form of increasing concentrations of H2O2. H2O2 at concentrations ranging from 5 nM to 5 mM produced a statistically signif-

DISCUSSION We report that estrogen at a physiological concentration influences cholinergic function of basal forebrain neurons in vitro by increasing HACU and newly synthesized ACh. As both changes occur without a concomitant increase in ChAT activity, it is reasonable to assume the increase in ACh results from the change in HACU consistent with the role for HACU as the rate-limiting step in ACh synthesis (Hebb et al., 1964; Bhatnagar and MacIntosh, 1967; Guyenet et al., 1973; Zapata et al., 2000). The rapid ERK 1/2 phosphorylation and blockade of estrogen-induced changes in HACU by the MEK inhibitor, U0126, implicate phosphorylated ERK 1/2 in the initial signaling response mediating prolonged changes in HACU and ACh. Furthermore, estrogen enhancement of HACU does not result from increased neuronal survival of basal forebrain neurons in culture. The ability of ICI 182,780 to block E2mediated changes in HACU suggests that it is selective and not due to non-specific steroid effects known to influence lipid and protein membrane composition (Whiting et al., 2000; Sommer and Crowley, 2001). Zapata et al. (2000) have observed changes in LACU as a result of membrane phospholipid depletion. In our studies LACU remains unaltered reinforcing the idea that the changes in the estrogen-mediated changes in HACU are not the result of a general influence on membrane integrity. Thus, our data suggest estrogen supports cholinergic function through the rapid activation of a kinase cascade rather than a genomic or non-specific mechanism. Work in understanding how estrogen acts on cognition has concentrated on the estrogen-mediated changes in hippocampal spine density and synaptogenesis favoring a NMDA-dependent mechanism of GABAergic disinhibition of pyramidal neurons mediated through a non-genomic mechanism such as altered dendritic phosphoAkt (McEwen et al., 2001b for review). Very recent reports, however, indicate that septal cholinergic inputs are required for the estrogen-mediated enhancement of working memory and NMDA receptor binding in the CA1 region of the hippocampus (Daniel and Dohanich, 2001) and estradiol-mediated changes in hippocampal spine density in rats (Leranth et al., 2000; Lam and Leranth, 2003: Rudick et al., 2003). These findings together with the detection of ER-␣ within septal cholinergic terminals in the hippocampus (Towart et al., 2003) argue for the possibility of alternative mechanisms of estrogen, potentially non-genomic, in contributing to cognition. Earlier studies of ovariectomized rats reported 24 –50% reductions in HACU in corti-

J. L. Pongrac et al. / Neuroscience 124 (2004) 809 – 816

B.

17 β -Estradiol

______________ 20

30

_ _

3

a,b

2

1

0

0/0

5/0

5/10

17β -estradiol (nmol)/U0126 (µmol)

_ _

C.

1500 a,b

(dpm/5min)

ERK1 ERK2

10

ACh Synthesized

pERK1 pERK2

0

HACU

Time (min)

(nmol/5 min/mg protein)

A.

813

1000

500

0

0/0

0/5

10/5

17β -estradiol (nmol)/U0126 (µmol) Fig. 3. Phosphoryation of ERK 1/2 by estrogen. (A) Immunoblots derived from Western blot analyses are shown demonstrating ERK 1/2 phosphorylation at the indicated time points following administration of 5 nM estradiol at 4 DIV to rat basal forebrain neurons in the top panel with total ERK 1/2 shown from the same blot that was stripped and reprobed in the bottom panel. The change in the proportion of phosphorylated ERK 1/2 relative to total ERK 1/2 in response to estrogen at a dose of 5 nM is suggestive of an increase in ERK 1/2 activity. Representative blot from n⫽3 independent cultures. Images were normalized for brightness. The reversible MEK 1/2 inhibitor administered for 25 min following estrogen blocked the estrogen-mediated increase in (B) HACU (mean⫾S.E.M. [3H]choline expressed as nmoles/5 min/mg of protein; control, 1.196⫾0.03; estradiol, 2.271⫾0.04; estradiol⫹U0126, 1.169⫾0.03; n⫽5) and (C) newly synthesized ACh (mean⫾S.E.M. [3H]ACh expressed as d.p.m./5 min; control, 664.3⫾162; estradiol, 804.6⫾215; estradiol⫾U0126, 587.9⫾183) when assayed on 8 DIV. a Significantly different from control or estradiol⫹U0126 (Neuman-Keuls multiple comparison, P⬍0.05).

cal and hippocampal tissues (O’Malley et al., 1987; Singh et al., 1994). Estrogen replacement from 5 days to 2 weeks restores HACU to or beyond control levels (O’Malley et al., 1987; Singh et al., 1994; Gibbs, 2000). Singh et al. (1994)

Cytotoxicity (%)

100

-17 β -estradiol +17 β -estradiol

75

50 25 0

0

5

50

500

5000

H2O2 (nM) Fig. 4. Effect of estrogen on H2O2-induced LDH release. Rat basal forebrain neurons were treated with 5 nM estradiol at 4 DIV and survival was measured by release and detection of LDH activity in the culture medium at 8 DIV. Even in the presence of an oxidant stressor, H2O2, that produced a dose-dependent cytotoxicity (80% cell death, P⬍0.0001, two-way ANOVA), 5 nM estrogen was unable to protect neurons (P⫽0.5344, two-way ANOVA; n⫽4 independent cultures). This suggests that the estrogen-mediated increase in HACU was not the result of differential viability between vehicle- and estrogen-treated neurons.

suggested the improvements in cognitive function under these conditions result from changes in HACU, but assessing this is difficult given that changes in ChAT activity also occur and have similar potential to alter ACh synthesis and cognition (Luine, 1985; Singh et al., 1994; Gibbs, 2000). At 5 weeks following estrogen depletion, however, Singh et al. (1994) found decreases in HACU without changes in ChAT activity in cortex establishing a differential regulation of these two cholinergic markers by estrogen. Kar et al. (1998) have also shown a preferential vulnerability of BFCNs to amyloid-␤ compared with striatal cholinergic neurons resulting in a loss of HACU and ACh release without changes in ChAT activity. Our data from estrogen-treated basal forebrain neuron cultures substantiate this notion that HACU and ChAT activity are regulated through distinct mechanisms and that estrogen influences HACU and ACh through a MAP kinase signaling mechanism. This makes our model appropriate for further exploration of estrogenic effects on HACU-dependent ACh changes. Ultrastructural analyses of ER expression has revealed ER-␣ is colocalized with the vesicular ACh transporter and discretely arranged in axons and axon terminals that synapse with the hippocampus in addition to nuclear ER (Towart et al., 2003). Estrogen actions could then be mediated directly in the synapse where HC-3-sensitive choline trans-

814

J. L. Pongrac et al. / Neuroscience 124 (2004) 809 – 816

porters are located in target tissue (Misawa et al., 2001). Such an arrangement in the synapse would be favorable for non-genomic mechanisms of action for estrogen through a rapid second messenger response. Although, we did not observe changes in phosphoAkt using Western blot techniques (data not shown), our data are consistent with non-genomic estrogen signaling as determined by the ERK 1/2-dependence of the estrogen-mediated change in HACU and ACh. Vasudevan et al. (2001), however, demonstrate an ER-␣-dependent synergistic induction of an ERE-responsive reporter gene as a result of a rapid membrane-dependent estrogen signal preceding a slower membrane-independent estrogen signal in a human neuroblastoma cell line. Thus, the contribution of transcriptional bi-products to estrogen signaling should not be ignored. Glutathione peroxidase and metallothionein are induced by estrogen (Bednarek-Tupikowska et al., 2001; Akova et al., 2001) and may act together as thioreductants to protect HACU presumably against cysteine oxidation (Guermonprez et al., 2001) without necessarily affecting cell survival. Other transcriptional products of estrogen activity in cholinergic neuron or neuron-like models include the neurotrophins and their receptors (Gibbs et al., 1994a; Sohrabji et al., 1994). Our results are in direct contrast to those of the effects of NGF in primary cultures of basal forebrain neurons grown under the same conditions. That is, NGF elevates ChAT activity, but not HACU (Pongrac and Rylett, 1996; Auld et al., 2001) suggesting the estrogen-mediated increases in HACU observed in the present study are not mediated through neurotrophins. In contrast the observed changes in tyrosine kinase A and ChAT activity in the hippocampus of ovariectomized estrogenreplaced rats (Gibbs et al., 1994a) may depend on estrogen signaling mechanisms in cell-types not represented in our basal forebrain cultures such as glia or hippocampalderived cells. For instance, ACh release induces NGF expression in the hippocampus (Knipper et al., 1994) which in turn can potentially increase tyrosine kinaseA (Gibbs and Pfaff, 1994) and ChAT activity (Williams and Rylett, 1990). Thus, estrogen may act through enhancement of the cholinergic function in an intact system to augment NGF signaling and ChAT activity. Finally, estrogen might also regulate the amount of high-affinity choline transporter which should be explored in future studies. Our data also suggest that the effect of estradiol on HACU was not due to selective survival of the cholinergic neurons through an anti-oxidant mechanism of estrogen as proposed by several researchers (Green et al., 1997; Moosmann and Behl, 1999; Bae et al., 2000; Behl, 2000; Harms et al., 2000). In the present study, estrogen at a dose of 5 nM did not protect cultures against high doses of H2O2. This is in contrast to the protection reported by Brinton et al. (2000) conferred by conjugated equine estrogens following exposure of BFCN cultures to 20 ␮M H2O2. Due to the presence of progestins, androgens and uncharacterized molecules in the formulation, however, it is very difficult to assess whether the improvement in viability resulted from the specific influence of estrogen. Additionally, the cultures were not characterized for func-

tional competence of BFCNs. In conclusion, our data demonstrate an ERK-dependent mechanism for estrogen in the regulation of the cholinergic phenotype independent of an improvement in survival or influence from neurotrophins. Acknowledgements—This work was supported by the National Institute of Health Grants NS38319 (D.B.D.) and NS28896 (R.B.G.), and University of Pittsburgh Alzheimer’s Disease Research Center Grant AG005133 (D.B.D.). J.L.P. was also supported by a Postdoctoral Fellowship for the Natural Sciences and Engineering Council of Canada (NSERC).

REFERENCES Akova B, Surmen-Gur E, Gur H, Dirican M, Sarandol E, Kucukoglu S (2001) Exercise-induced oxidative stress and muscle performance in healthy women: role of vitamin E supplementation and endogenous oestradiol. Eur J Appl Physiol 84:141–147. Auld DS, Mennicken F, Day JC, Quirion R (2001) Neurotrophins differentially enhance acetylcholine release, acetylcholine content and choline acetyltransferase activity in basal forebrain neurons. J Neurochem 77:253–262. Bae YH, Hwang JY, Kim YH, Koh JY (2000) Anti-oxidative neuroprotection by estrogens in mouse cortical cultures. J Korean Med Sci 15:327–336. Bednarek-Tupikowska G, Bohdanowicz-Pawlak A, Bidzinska B, Milewicz A, Antonowicz-Juchniewicz J, Andrzejak R (2001) Serum lipid peroxide levels and erythrocyte glutathione peroxidase and superoxide dismutase activity in premenopausal and postmenopausal women. Gynecol Endocrinol 15:298 –303. Behl C (2000) Vitamin E protects neurons against oxidative cell death in vitro more effectively than 17-beta estradiol and induces the activity of the transcription factor NF-kappaB. J Neural Transm 107:393–407. Beyer C, Raab H (1998) Nongenomic effects of oestrogen: embryonic mouse midbrain neurons respond with a rapid release of calcium from intracellular stores. Eur J Neurosci 10:255–262. Bhatnagar SP, MacIntosh FC (1967) Effects of quaternary bases and inorganic cations on acetylcholine synthesis in nervous tissue. Can J Physiol Pharm 45:249 –268. Brinton RD, Chen S, Montoya M, Hsieh D, Minaya J, Kim J, Chu H-P (2000) The women’s health initiative estrogen replacement therapy is neurotrophic and neuroprotective. Neurobiol Aging 21:475–496. Bottenstein JE, Sato GH (1976) Growth of a rat neuroblastoma cell line in serum-free supplemented medium. Proc Natl Acad Sci USA 76:514 –517. Daniel JM, Dohanich GP (2001) Acetylcholine mediates the estrogeninduced increase in NMDA receptor binding in CA1 of the hippocampus and the associated improvement in working memory. J Neurosci 21:6949 –6956. Fonnum F (1969) Radiochemical microassays for the determination of choline acetyltransferase and acetylcholinesterase activities. Biochem J 115:465–479. Gabor R, Nagle R, Johnson DA, Gibbs RB (2003) Estrogen enhances potassium-stimulated acetylcholine release in the rat hippocampus. Brain Res 962:244 –247. Gibbs RB, Hashash A, Johnson DA (1997) Effects of estrogen on potassium-stimulated acetylcholine release in the hippocampus and overlying cortex of adult rats. Brain Res 749:143–146. Gibbs RB, Wu D, Hersh LB, Pfaff DW (1994a) Effects of estrogen replacement on the relative levels of choline acetyltransferase, trkA, and nerve growth factor messenger RNAs in the basal forebrain and hippocampal formation of adult rats. Exp Neurol 129:70 – 80. Gibbs RB, Pfaff DW (1994) In situ hybridization detection of trkA

J. L. Pongrac et al. / Neuroscience 124 (2004) 809 – 816 mRNA in brain: distribution, colocalization with p75NGFR and upregulation by nerve growth factor. J Comp Neurol 341:324 –339. Gibbs RB, Pfaff DW (1992) Effects of estrogen and fimbrial/fornix transection on p75NGFR and ChAT expression in the medial septum and diagonal band of Broca. Exp Neurol 116:23–29. Gibbs RB (1997) Effects of estrogen on basal forebrain cholinergic neurons vary as a function of dose and duration of treatment. Brain Res 757:10 –16. Gibbs RB (2000) Effects of gonadal hormone replacement on measures of basal forebrain cholinergic function. Neuroscience 101: 931–938. Gibbs RB (2002) Basal forebrain cholinergic neurons are necessary for estrogen to enhance acquisition of a delayed matching-to-position T-maze task. Horm Behav 42:245–247. Green PS, Gordon K, Simpkins JW (1997) Phenolic A ring requirement for the neuroprotective effects of steroids. J Steroid Biochem Mol Biol 63:4 –6. Goldberg AM, McCaman RE (1973) The determination of picomole amounts of acetylcholine in mammalian brain. J Neurochem 20: 1–8. Guyenet P, Lefresne P, Rossier J, Beaujouan JC, Glowinski J (1973) Inhibition by hemicholinium-3 of (14C)-acetylcholine synthesis and (3H) choline high-affinity uptake in rat striatal synaptosomes. Mol Pharmacol 9:630 –639. Guermonprez L, Ducrocq C, Gaudry-Talarmain YM (2001) Inhibition of acetylcholine synthesis and tyrosine nitration induced by peroxynitrite are differentially prevented by antioxidants. Mol Pharmacol 60:838 –846. Harms C, Lautenschlager M, Bergk A, Katchanov J, Freyer D, Kapinya HU, Megow D, Dirnagl U, Weber JR, Hortnagl H (2000) Differential mechanisms of neuroprotection by 17 beta-estradiol in apoptotic versus necrotic neurodegeneration. J Neurosci 21:2600 –2609. Hebb CO, Ling GM, McGeer EG, McGeer PL, Perkins D (1964) Effect of locally applied hemicholinium on the acetylcholine content of the caudate nucleus. Nature (Lond) 204:1309 –1311. Ivanova T, Kuppers E, Engele J, Beyer C (2001) Estrogen stimulates brain-derived neurotrophic factor expression in embryonic mouse midbrain neurons through a membrane-mediated and calcium-dependent mechanism. J Neurosci Res 66:221–230. Jope RS (1979) High affinity choline transport and acetyl-CoA production in brain and their roles in the regulation of acetylcholine synthesis. Brain Res Rev 1:313–344. Kar S, Issa AM, Seto D, Auld DS, Collier B, Quirion R (1998) Amyloid beta-peptide inhibits high-affinity choline uptake and acetylcholine release in rat hippocampal slices. J Neurochem 70:2179 –2187. Knipper M, da Penha Berzaghi M, Blochl A, Breer H, Thoenen H, Lindholm D (1994) Positive feedback between acetylcholine and the neurotrophins nerve growth factor and brain-derived neurotrophic factor in the rat hippocampus. Eur J Neurosci 6:668 –671. Laemmeli UK (1970) Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227:680 –685. Lam T-T, Leranth C (2003) Role of the medial septum diagonal band of Broca cholinergic neurons in oestrogen-induced spine synapse formation on hippocampal CA1 pyramidal cells of female rats. Eur J Neurosci 17:1997–2005. Leranth C, Shanabrough M, Horvath TL (2000) Hormonal regulation of hippocampal spine synapse density involves subcortical mediation. Neuroscience 101:349 –356. Luine VN (1985) Estradiol increases choline acetyltransferase activity in specific basal forebrain nuclei and projection areas of female rats. Exp Neurol 89:484 –490. McEwen BS (1999) The molecular and neuroanatomical basis for estrogen effects in the central nervous system. J Clin Endocrinol Metab 84:1790 –1797. McEwen BS (2001a) Estrogens effects on the brain: multiple sites and molecular mechanisms. J Appl Physiol 91:2785–2801.

815

McEwen BS, Akama K, Alves S, Brake WG, Bulloch K, Lee S, Li C, Yuen G, Milner TA (2001b) Tracking the estrogen receptor in neurons: implications for estrogen-induced synapse formation. Proc Natl Acad Sci USA 98:7093–7100. Miettinen RA, Kalesnykas G, Koivisto EH (2002) Estimation of the total number of cholinergic neurons containing estrogen receptor-alpha in the rat basal forebrain. J Histochem Cytochem 50:891–902. Misawa H, Nakata K, Matsuura J, Nagao M, Okuda T, Haga T (2001) Distribution of the high-affinity choline transporter in the central nervous system of the rat. Neuroscience 105:87–89. Moosmann B, Behl C (1999) The antioxidant neuroprotective effects of estrogens and phenolic compounds are independent from their estrogenic properties. Proc Natl Acad Sci USA 96:8867–8872. O’Malley CA, Hautamaki RD, Kelley M, Meyer EM (1987) Effects of ovariectomy and estradiol benzoate on high affinity choline uptake, ACh synthesis, and release from rat cerebral cortical synaptosomes. Brain Res 403:389 –392. Pongrac JL, Rylett RJ (1996) Differential effects of nerve growth factor on expression of choline acetyltransferase and sodium-coupled choline transport in basal forebrain cholinergic neurons in culture. J Neurochem 66:804 –810. Pongrac JL, Rylett RJ (1998a) Optimization of serum-free culture conditions for growth of embryonic rat cholinergic basal forebrain neurons. J Neurosci Methods 84:69 –76. Rabbini O, Panickar KS, Rajakumar G, King MA, Bodor N, Meyer EM, Simpkins JW (1997) 17 beta-estradiol attenuates fimbrial lesioninduced decline of ChAT-immunoreactive neurons in the rat medial septum. Exp Neurol 146:179 –186. Rudick CN, Gibbs RB, Woolley CS (2003) A role for the basal forebrain cholingeric system in estrogen-induced disinhibition of hippocampal pyramidal cells. J Neurosci 23:4479 –4490. Singh M, Meyer EM, Millard WJ, Simpkins JW (1994) Ovarian steroid deprivation results in a reversible learning impairment and compromised cholinergic function in female Sprague-Dawley rats. Brain Res 644:305–312. Singh M, Setalo G, Guan X, Warren M, Toran-Allerand CD (1999) Estrogen-induced activation of mitogen-activated protein kinase in cerebral cortical explants: convergence of estrogen and neurotrophin signaling pathways. J Neurosci 19:1179 –1188. Shughrue PJ, Scrimo PJ, Merchenthaler I (2000) Estrogen binding and estrogen receptor characterization (ER␣ and ER␤) in the cholinergic neurons of the rat basal forebrain. Neuroscience 96:41–49. Smith YR, Minoshima S, Kuhl DE, Zubieta J-K (2001) Effects of long-term hormone therapy on cholinergic synaptic concentrations in healthy postmenopausal women. J Clin Endocrinol Metab 86: 679 –684. Sohrabji F, Miranda RC, Toran-Allerand CD (1994) Reciprocal regulation of estrogen and NGF-receptors by their ligands in PC12 cells. J Neurobiol 25:974 –988. Sommer P, Crowley HM (2001) Effect of steroid hormones on membrane profiles of HeLa S3 cells. Cell Biol Int 25:103–111. Toran-Allerand CD, Miranda RC, Bentham W, Sohrabji F, Brown TJ, Hochberg RB, MacLusky NJ (1992) Estrogen receptors co-localize with low-affinity nerve growth factor receptors in cholinergic neurons of the basal forebrain. Proc Natl Acad Sci USA 89:4668 –4672. Towart LA, Alves SE, Znamensky V, Hayashi S, McEwen BS, Milner TA (2003) Subcellular relationships between cholinergic terminals and estrogen receptor-alpha in the dorsal hippocampus. J Comp Neurol 463:390 –401. Vasudevan N, Kow L-M, Pfaff DW (2001) Early membrane estrogenic effects required for full expression of slower genomic actions in a nerve cell line. Proc Natl Acad Sci USA 98:12267–12271. Watters JJ, Campbell JS, Cunningham MJ, Krebs EG, Dorsa DM (1997) Rapid membrane effects of steroids in neuroblastoma cells: effects of estrogen on mitogen activated protein kinase signaling cascade and c-fos immediate early gene transcription. Endocrinology 138:4030 –4033.

816

J. L. Pongrac et al. / Neuroscience 124 (2004) 809 – 816

Whiting KP, Restall CJ, Brain PF (2000) Steroid hormone-induced effects on membrane fluidity and their potential roles in nongenomic mechanisms. Life Sci 67:743–757. Williams LR, Rylett RJ (1990) Exogenous nerve growth factor increases the activity of high-affinity choline uptake and choline

acetyltransferase in brain of Fisher 344 male rats. J Neurochem 55:1042–1049. Zapata A, Capdevila JL, Trullas R (2000) Role of high-affinity choline uptake on extracellular choline and acetylcholine evoked by NMDA. Synapse 35:272–280.

(Accepted 13 January 2004)