Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2010 Jan 23.
Published in final edited form as: Neuroscience. 2008 Oct 10;158(2):602–609. doi: 10.1016/j.neuroscience.2008.10.003

Soy phytoestrogens are neuroprotective against stroke-like injury in vitro

Derek A Schreihofer 1, Lori Redmond 2
PMCID: PMC2652887  NIHMSID: NIHMS94262  PMID: 18976694

Abstract

Diets high in soy are neuroprotective in experimental stroke. This protective effect is hypothesized to be mediated by phytoestrogens contained in soy, because some of these compounds have neuroprotective effects in in vitro models of cell death. We tested the ability of the soy phytoestrogens genistein, daidzein, and the daidzein metabolite equol to protect primary cortical neurons from ischemic-like injury in vitro at doses typical of circulating concentrations in human populations (0.1-1 μM). All three phytoestrogens inhibited LDH release from cells exposed to glutamate toxicity or the calcium-ATPase inhibitor, thapsigargin. In cells exposed to hypoxia or oxygen-glucose deprivation (OGD), pretreatement with the phytoestrogens inhibited cell death in an estrogen receptor (ER) dependent manner. Although OGD results in multiple modes of cell death, examination of α-spectrin cleavage and caspase-3 activation revealed that the phyoestrogens were able to inhibit apoptotic cell death in this model. In addition, blockade of phosphoinositide 3-kinase prevented the protective effects of genistein and daidzein, and blockade of mitogen-activated protein kinase prevented genistein-dependent neuroprotection. These results suggest that pretreatment with dietary levels of soy phytoestrogens can mimic neuroptotective effects observed with estrogen and appear to use the same ER-kinase pathways to inhibit apoptotic cell death.

Introduction

In vivo and in vitro studies have demonstrated powerful neuroprotective effects of estrogens against a variety of insults. In rodents, estrogen reduces injury caused by focal cerebral ischemia (Gibson et al., 2006), global ischemia (Jover et al., 2002), cerebral trauma (Bramlett and Dietrich, 2001), β-amyloid (Aguado-Llera et al., 2006), and glutamate (Mendelowitsch et al., 2001) or MPTP toxicity (Callier et al., 2001, Morissette et al., 2007). In vitro, similar neuroprotectective effects have been demonstrated in primary cortical (Singer et al., 1999, Honda et al., 2000, Sribnick et al., 2004), hippocampal (Nilsen and Brinton, 2002, Cimarosti et al., 2005, Chen et al., 2006) and mesencephalic cultures (Callier et al., 2001). Furthermore, these neuroprotective effects appear to utilize multiple mechanisms depending on the injury and cell type. These mechanisms include classical transcriptional signaling through estrogen receptors alpha and beta, activation of multiple kinases such as mitogen activated protein kinase (MAPK) and phosphoinositide 3-knase (PI3K), modulation of intracellular calcium levels, and anti inflammatory effects (for review see (Amantea et al., 2005). Unfortunately, these well-demonstrated neuroprotective effects of estrogen may not be applicable to postmenopausal women or men because of the potential deleterious side-effects of hormone replacement therapy, such as increased cancer and stroke risk (Wassertheil-Smoller et al., 2003, Brass, 2004).

Recent advances in pharmaceuticals have resulted in the development of selective estrogen receptor modulators (SERMs) that may provide many of the beneficial effects of estrogen without the deleterious side effects. However, many women now consider the use of complementary and alternative therapies, including the use of high soy diets or soy isoflavone supplements for menopausal symptoms (Wathen, 2006). It is increasingly clear that physiologically attainable doses of isoflavones, which can behave as phytoestrogens, can mimic the some of the neuroprotective effects of estrogens. For example, a high soy diet reduces stroke injury in female and male rats (Schreihofer et al., 2005, Burguete et al., 2006) and the soy isoflavone genistein is neuroprotective in a mouse cerebral ischemia model (Trieu and Uckun, 1999). In vitro some soy isoflavones can protect primary neurons from glutamate toxicity (Zhao et al., 2002), thapsigargin-induced apoptosis (Linford and Dorsa, 2002), and β-amyloid toxicity (Zeng et al., 2004). However, the mechanisms underlying protection from ischemic injury remains unclear. In this study we hypothesized that soy phytoestrogens would be neuroprotective in an in vitro model of ischemic stroke and that the mechanism would involve estrogen receptor activation.

Experimental procedures

Primary Cortical Culture

Cortical neurons were cultured from Long Evans rats as previously described (Redmond, et al. 2002). Briefly, embryonic day 18 cortices were digested with papain (Worthington), dissociated, and plated onto polylysine and laminin coated wells in 24 well plates at 1.86 × 105 cells/cm2. Cultures were maintained in Neurobasal media supplemented with 2% B27, 1mM glutamine and penicillin/streptomycin (Invitrogen). Cultures were maintained at 37C in 5% CO2 and media was changed at 4 and 7 days in vitro (DIV). Estrogen receptor (ER) expression in embryonic cortical neurons was determined by real time RT-PCR, immunoblotting, and immunohistochemistry as described below. For mRNA comparison, pooled adult rat cortical, hippocampal, and hypothalamic tissue was used. For immunoblotting, rat uterus and mouse αT3 pituitary cells were used.

Apoptotic challenges

In all experiments control and challenged wells underwent the same number of washes and medium changes. Cortical cells were used at 7 DIV. Cells were pretreated with estrogens or phytoestrogens for 24 hours in a fresh change of NB/B27, and treatments were present throughout the experiment. Phytoestrogens (genistein, daidzein, and equol) were obtained from LC Laboratories (Woburn, MA). Compounds were all dissolved and diluted in 95% ethanol to maintain a constant ethanol concentration in all experiments. The following day, one of the following challenges was applied and cell viability was assessed at 24 and 48 hours. For excitotoxicity, cells were challenged with 300 μM glutamate in Hanks Buffered Salt Solution (HBSS) for 10 minutes and then replaced with conditioned medium. Thapsigargin (50 nM) was used to induce apoptosis by increasing intracellular calcium levels. Cells were treated with thapsigargin in NB/B27 for 48h in the presence of treatments. Mitochondrial respiration was inhibited with KCN (1 mM) treatment. Cells were exposed to KCN in NB/B27 for 2 hours and then replaced with conditioned medium. The addition of 2 mM 2-deoxy-D-glucose to KCN treatment was also examined as a model for ischemia.

Hypoxia was induced by placing uncovered plates in a prewarmed and humidified chamber (Billups-Rothenberg) that was subsequently flushed with 95%N2/5%CO2 for 15 minutes. An oxygen sensor placed in the chamber confirmed oxygen levels of less than 0.6%. The sealed chamber was then placed in a standard incubator for 18 hours. Plates were then removed from the chamber and allowed to re-oxygenate in the tissue culture incubator for 6 hours before assessment.

Oxygen-glucose deprivation (OGD) was achieved using the same hypoxic chamber. After 24 hours of pretreatment in 500 μl of NB/B27, cells were washed twice with HBSS and then placed in glucose-free HBSS (300 μl) containing treatments and subjected to 2 or 5 hours of hypoxia, as described above. Following hypoxia, cells were allowed to reoxygenate in fresh Neurobasal medium containing 0.5 mM glutamine, but no B27 supplement. Control wells were subjected to the same washes but were maintained in HBSS+ 25 mM glucose under normoxic conditions during the OGD period.

Mechanism of action

The mechanism of phytoestrogen dependent neuroprotection was assessed with specific inhibitors of potential protective signaling molecules. Dependence on ER signaling was assessed with pretreatment (-30 min) with the ERα/β antagonist ICI182,780 (Tocris Cookson Inc., Ellisville, MO). Dependence on PI3K activated pathways was determined by pretreatment with the PI3K inhibitor LY294002 (Tocris Cookson), and dependence on MAPK activated pathways was determined by pretreatment with the specific MAPK kinase inhibitor PD98059 (Tocris Cookson).

Assessment of cell death

Cell death was determined 24 and 48 hours after challenge by the release of lactate dehydrogenase (LDH), a cytoplasmic enzyme released from cells with compromised cell membranes, and a marker of both apoptosis and necrosis. LDH release into the culture medium was detected using a colormetric reaction read at an absorbance at 490 nm on a Tecan Genios plate reader using a commercial kit (Takara). Maximal LDH release was determined by treating control wells for 60 minutes with 1% triton-X 100 to lyse all cells.

Apoptotic cell death was examined by in situ detection of caspase-3 activity using a fluorogenic substrate (FAD-VMK-FITC) from Promega. Active caspase-3 cleaves the substrate, resulting in green fluorescence. Images from treated wells were captured on an Olympus IX50 inverted microscope with an Optronics Magnifier SP digital camera. Cleavage of α-spectrin and caspase-3 were also determined by immunoblotting, as described below. Calcein AM (Molecular Probes), a viability marker that is hydrolyzed by intracellular esterases in living cells, was used in some cultures to confirm results from apoptotic assays.

Real-time RT-PCR

Total RNA was collected and isolated using a commercial kit (Qiagen RNEasy). RNA was reverse transcribed with oligo-deoxy-thymidine primers using Omniscript (Qiagen). Real time PCR was performed using ERα, ERβ and GPR30 predesigned Taq-Man assays from Applied Biosystems. GAPDH was amplified as an endogenous control gene. RT-PCR was performed two independent times from independent samples. Relative message levels were normalized to expression in the adult rat cerebral cortex.

Immunoblotting

Total protein was collected in MPER lysis buffer (Pierce) containing a protease inhibitor cocktail (HALT, Pierce) from cortical cells grown in 35 mm plates. Subcellular fractionation of the cytoplasm and nucleus was achieved with the NPER lysis kit (Pierce). Protein concentrations were determined in duplicate using the Pierce BCA protein assay reagent and bovine serum albumin to generate standard curves. Absorbance at 595 nm was detected on a Tecan Genios plate reader. Proteins were aliquoted and frozen at −80C to avoid multiple freeze-thaw cycles. Equivalent amounts of protein were separated by SDS-polyacrylamide gels electrophoresis using Pierce Precise precast gels. Proteins were transferred to nitrocellulose (BioRad) and detected with the following antibodies: ERα (LabVision Ab15, 1:500); ERβ (Upstate 06-029, 4 μg/ml); α-spectrin (Chemicon MAB1622, 1:1000); active caspase-3 (Cell Signaling Technologies 9661, 1:1000); β-actin (Sigma AC15, 1:1000). Secondary detection was achieved with LiCor infrared-labeled antibodies (1:10,000) and the LiCor Odyssey infrared scanner. Expression levels were determined using Odyssey software and normalized to blot background levels and β-actin.

Immunocytochemistry

For immunocytochemical detection of ERs, cells grown in chamber slides (Falcon) were washed with PBS and fixed for 20 minutes at room temperature with 4% formaldehyde. Cells were permeablized with 0.2% triton X-100 in PBS for 10 min at 4°C. Non-specific binding was blocked with 5% BSA and 1% horse serum in PBS for 1 h at 4°C, and cells were incubated with primary antibodies overnight at 4C. ERα was detected with Upstate ERα (C1355, 1:5,000) and ERβ was detected with Upstate ERβ (06-029, 1:100). Secondary detection was accomplished with Cy3 conjugated donkey anti-rabbit antibody (Jackson Immunologicals, 1:500), and nuclei were counterstained with DAPI (Pierce). The slides were coverslipped with Vectashield (Vector Laboratories), and images were captured with an Optronics Magnifier SP digital camera on an Olympus BX60 microscope with appropriate fluorescent filters.

Statistics

All data was collected from at least 4 independent experiments. Data are all presented as normalized to vehicle controls, but statistical analyses were performed on raw data. Differences among groups were detected by analysis of variance for repeated measures and pair-wise comparisons to vehicle control wells were made with Dunnet's test using InStat software (GraphPad). Data presented in Figure 5 and 6 did not fit a normal distribution and was analyzed by non-parametric methods using Friedman's test followed by Dunn's post-hoc comparisons. In all cases a P value of <0.05 was considered significant.

Figure 5.

Figure 5

Open in a new tab

Activation of caspase-3 in OGD exposed cells. A. Representative western blot showing p17 caspase-3 cleavage product in 1 μM phytoestrogen treated cells exposed to OGD. B. Quantification of western blot band density for p17 kDa fragment of caspase 3. Values represent means of 5 independent experiments ± SE. Optical density of p17 was normalized to β-actin. Asterisk (*) represents significant difference (P<0.05) from vehicle (non-OGD) control (Friedman Statistic 11.328, P=0.0231, Dunn post-hoc P<0.05). C. Representative photomicrographs from of cortical cells stained with the in situ caspase 3 maker VAD-FMK-FITC 48 hours after OGD.

Figure 6.

Figure 6

Open in a new tab

Role of PI3K and MAPK in phytoestrogen neuroprotection. LDH release in cortical cells pretreated 24 h with 1 μM genistein (Gen), daidzein (Daid), or equol alone or + PD98059 (50 μM, solid bars) or LY294002 (10 μM, hatched bars) and subjected to 2 hours of OGD and 48 hours recovery. Values are normalized to the vehicle OGD only condition (OGD) and represent mean ± SE of 6 independent experiments (Friedman Statistic 9, P=0.0218, Dunn post-hoc P<0.05). Asterisks (*) represent significant difference (P<0.05) from vehicle control.

Results

Because the reported expression of ER expression in primary cortical cultures has been inconsistent across the literature, we first examined the expression of ER in primary cortical neurons used in our experiments. After 7 DIV, approximately 95% of cells were neurons, as detected by MAP2 and GFAP immunostaining (not shown). Real time RT-PCR revealed expression of ERα levels 9-fold greater than the adult rat cortex in primary cortical neurons under our culture conditions (Figure 1). ERβ mRNA levels were 40% of those in the adult cortex, and GPR30 levels were only 11% of cortical levels (Figure 1). For comparison, hypothalamic levels of ER mRNA are 20-50 times higher than in cortex (Figure 1). Immunoblotting revealed ERα protein in the cultured cortical neurons was primarily nuclear (Figure 1B) and ERβ was primarily cytoplasmic in our cells (Figure 1B), an observation confirmed by immunostaining (Figure 1C). These analyses were performed twice with similar results, but are not intended to be quantitative.

Figure 1.

Figure 1

Open in a new tab

Estrogen receptor expression in cultured embryonic neurons. A. Steady state mRNA levels of ERα, ERβ, and GPR30 in cultured embryonic cortical cells, adult rat cortex, hypothalamus, and hippocampus determined with real-time RT-PCR. B. Protein levels of ERα and ERβ in cultured cortical whole cell (WC), cytoplasm (Cyt), and nuclear (Nuc) extracts compared to adult rat uterus (Ut) and mouse αT3 pituitary cells determined with immunoblotting. C. ERα and ERβ staining and DAPI nuclear localization in cultured cortical cells.

Estradiol is neuroprotective in primary cortical neurons faced with several different apoptotic challenges. Similarly, various phytoestrogens can be neuroprotective. We compared the neuroprotective efficacy of physiological concentrations of 17β-estradiol (E2) to concentrations of the soy isoflavones genistein and daidzein (0.1-1 μM) at levels observed in the plasma of individuals on a high soy diet (Setchell, 2001). In addition, we examined the neuroprotective potential of the daidzein metabolite equol, which makes up the majority of circulating isoflavone in rodents. In agreement with published reports, E2 significantly reduced neuronal death in response to glutamate and thapsigargin (Figure 2). Similarly, pretreatment with the soy isoflavones genistein and daidzein also reduced LDH release under these conditions (Figure 2). Equol (1 μM) had no significant effect (Figure 2). Anoxia induced with KCN or oxygen-glucose deprivation induced by KCN combined with 2-deoxy-D-glucose led to LDH release that was not significantly reduced by E2 or phytoestrogen treatment (data not shown). Exposure of cortical cells to 18 hours of hypoxia in the presence of treatments led to increased LDH release (Figure 3). Pretreatment for 24 hours with 1 nM E2 or 0.1 and 1 μM soy isoflavones reduced LDH release in response to hypoxia (Figure 3). The ER antagonist ICI182,780 partially reversed the neuroprotective effects of genistein and daidzein, but not equol (Figure 3).

Figure 2.

Figure 2

Open in a new tab

Estrogen and phytoestrogens reduce LDH release in glutamate or thapsigargin treated cortical cells. A. LDH release in cortical cells pretreated 24 h with estradiol (E2, 10 nM), genistein (Gen, 1 μM), daidzein (Daid, 1 μM), or equol (1 μM) and subjected to 10 min of 300 mM glutamate followed by 48 hours of recovery. Values are normalized to the glutamate only condition (Veh) and represent mean ± SE of 5 independent experiments (F4,16=3.856, P=0.223). B. LDH release in cortical cells pretreated as in (A) and subjected to 48 hours of 50 nM thapsigargin. Values are normalized to the thapsigargin only condition (Veh) and represent mean ± SE of 5 independent experiments (F4,16=8.512, P=0.0007). Asterisks (*) represent significant difference (P<0.05) from vehicle control.

Figure 3.

Figure 3

Open in a new tab

Estrogen and phytoestrogens reduce LDH release in cortical cells exposed to hypoxia (Hypox) or OGD. A. LDH release in cortical cells pretreated 24 h with estradiol (E2, 10 nM), or 0.1 (open bar) or 1 μM (hatched bar) genistein, daidzein, or equol and subjected to 18 hours of hypoxia and 6 hours of reoxygenation. Filled bars represent 1 μM phytoestrogen treatments + 1 μM ICI182,780. Values are normalized to the vehicle only condition (Vehicle) and represent mean ± SE of 5-7 independent experiments. Asterisks (*) represent significant difference (P<0.05) from vehicle control (F10,40=2.082, P=0.0495). B. LDH release in cortical cells pretreated as in (A) and subjected to 90 minutes of hypoxia in glucose-free HBSS and 48 hours of reoxygenation in conditioned medium. Values are normalized to the vehicle only condition (Vehicle) and represent mean ± SE of 5-7 independent experiments. Asterisks (*) represent significant difference (P<0.05) from vehicle control (F10,40=2.486, P=0.0266). C. Representative photomicrographs of cortical cells stained with the viability marker calcein AM 48 hours after OGD.

Although glutamate, thapsigargin, and hypoxia in serum-free conditions mimic many of the characteristics of in vivo cerebral ischemia, glucose deprivation is also an important in vivo consequence of ischemia. Oxygen-glucose deprivation (OGD) for 2 hours leads to both apoptotic and necrotic cell death (Jones et al., 2004). Twenty-four hours of pretreatment with 10 nM E2, or 1 μM, but not 0.1 μM, genistein, daidzein, or equol significantly reduced LDH release 48 hours after OGD (Figure 3). Pretreatment with the ER antagonist ICI182,780 prevented the reduction of LDH, suggesting an ER-mediated effect (Figure 3). Cell staining with the viability marker calcein AM confirmed the reduction in cell viability in OGD cells and protection by phytoestrogens (Figure 3). Examination of α-spectrin cleavage revealed that OGD slightly, but significantly, increased the caspase-dependent 120 kDa fragment, but not the 150 kDa or calpain-dependent 145 kDa fragment (Figure 4). Phytoestrogens and estradiol treatment prevented accumulation of the 120 kDa fragment, but this effect was small (Figure 4). Similarly, phytoestrogens prevented the OGD-induced increase in caspase-3 cleavage (Figure 5). In situ staining for caspase activation confirmed results of immunoblotting (Figure 5).

Figure 4.

Figure 4

Open in a new tab

Cleavage of α-spectrin in cortical cells exposed to OGD. A. Representative western blot showing α-spectrin cleavage 48 hours after 2 hour OGD exposure in the absence or presence of 1 μM isoflavones or 10 nM estradiol. Quantification of western blot band density for p120 (B), p145 (C), and p150 (D) kDa fragments of α-spectrin. Values represent means of 5-7 independent experiments ± SE. Optical density of each fragment was normalized to the full-length α-spectrin and β-actin. Asterisk (*) represents significant difference (P<0.05) from vehicle (non-OGD) control (F5,20=3.566, P=0.0182 for B).

The neuroprotective effects of estrogen in primary cortical cultures have been attributed, in part, to their ability to activate the PI3 kinase (Honda et al., 2000, Honda et al., 2001, Wilson et al., 2002) and MAP kinase pathways (Singer et al., 1999, Mize et al., 2003). To determine whether the protective effects during OGD are mediated by these pathways, we pretreated cells with the specific PI3 kinase inhibitor LY294008 or the MEK inhibitor PD98059 just prior to estrogen or isoflavone treatment and OGD. Inhibition of both MEK and PI3K prevented genistein-induced protection from OGD (Figure 6) whereas only PI3K inhibition prevented daidzein-induced neuroprotection (Figure 6). Under our experimental conditions, the DMSO vehicle alone increased basal LDH release and prevented equol-dependent neuroprotection (Figure 6). Nevertheless, cotreatment with LY294008 increased LDH release in equol-treated cells exposed to OGD (Figure 6).

Discussion

Estradiol and estrogen-like compounds are powerful neuroprotective agents against numerous in vivo and in vitro apoptotic stimuli including stroke (Hurn and Brass, 2003, McCullough and Hurn, 2003, Alonso de Lecinana and Egido, 2006, Gibson et al., 2006). Diets high in soy leads to significant neuroprotection from focal cerebral ischemia in the rat (Schreihofer et al., 2005, Burguete et al., 2006). In addition, a major soy phytoestrogen, the isoflavone genistein, can reduce cerebral ischemic damage in mice (Trieu and Uckun, 1999). However, the precise mechanisms underlying these protective effects are unclear. In the present study, we show that dietary levels of the isoflavones genistein, daidzein, and the daidzein metabolite equol, can reduce cell death in cultured embryonic cortical cells exposed to challenges that mimic many of the events in cerebral ischemia including glutamate excitotoxicity, calcium mobilization, hypoxia, and oxygen-glucose deprivation (OGD). Estrogen receptors play a critical role in the protective effects of phytoestrogens as an ER antagonist blocked neuroprotection. In addition, activation of kinase survival pathways appears to play a role in protection by both genistein and daidzein, with the PI3K pathway appearing most important.

The complexity of in vivo studies makes elucidation of mechanisms of estrogen neuroprotection during cerebral ischemia difficult to study. Even with the availability of selective ligands and knockout mice controversy remains regarding the necessity for specific ERs in the protective effects of E2 (Dubal et al., 2000, Sampei et al., 2000, Carswell et al., 2004). Furthermore, the ability of estrogen to modulate cerebral blood flow (Pelligrino et al., 1998, Krause et al., 2006), inflammatory responses (Suzuki et al., 2007), blood brain barrier permeability (Bake and Sohrabji, 2004, O'Donnell et al., 2006), and glial cell function (Mahesh et al., 2006) makes dissecting out mechanisms difficult. Thus, in vitro models of injury have been employed as surrogates for the in vivo condition. Using embryonic cortical cells enriched for neurons we demonstrate that low micromolar concentrations of soy phytoestrogens can be directly neuroprotective against many of the insults associated with ischemic injury. However, the OGD model results in both apoptosis and necrosis (Jones et al., 2004) and although our results demonstrate inhibition of caspase pathways, it is likely that necrotic mechanisms are influenced as well.

Previous investigators have demonstrated neuroprotective effects of gensitein against glutamate (Singer et al., 1999, Sribnick et al., 2004) and thapsigargin-induced (Linford and Dorsa, 2002) apoptosis in embryonic cortical cells, and we report similar effects here with E2 and genistein. In addition we observed protective effects of daidzein. However, we saw no significant protection against chemical anoxia with KCN, chemical anoxia and hypoglycemia (KCN + 2-deoxy-D-glucose). These results contrast with those shown in organotypic cortical explant cultures treated for 1-7 days with E2, in which KCN/2-DG induced apoptosis was reduced 24-72 hours after challenge via an ER-dependent mechanism (Wilson et al., 2000). Although we observed a trend for neuroprotection by E2 and genistein after 48 hours, this effect failed to reach significance (data not shown).

In contrast to results with chemical ischemia, E2 and soy phytoestrogens all reduced LDH release in response to hypoxia. A protective effect of E2 has previously been shown in primary hippocampal neurons exposed to 15 hours of hypoxia, and this effect was attributed to non-genomic effects since it could be mimicked with E2-BSA and was not blocked with ICI182,780 (Heyer et al., 2005). We did observe diminution of the protective effects of genistein and daidzein with ICI182,780 suggesting the involvement of ERs. However, ICI182,780 did not alter the ability of equol to inhibit hypoxia-induced cell death. Thus, like estrogens, multiple mechanisms may be employed by different phytoestrogens.

Similar to hypoxia, neuronal death induced by OGD was significantly inhibited by E2 genistein, daidzein, and equol in an ER- and dose-dependent manner. However, although caspase activation and caspase-dependent α-spectrin cleavage were reduced by phytoestrogens, caspase activation was not extensive. Examination of caspase activation in cells confirmed immunoblotting. However, the total number of live cells was also reduced suggesting non-apoptotic cell death was also present.

Although soy phytoestrogens are often thought of as ERβ selective ligands, the doses used in the present experiment (0.1 and 1 μM) can readily activate ERα-dependent transcription in neurons (Schreihofer, 2005). The relative expression levels of ERs in our cells further supports a role for ERα-dependent neuroprotection. Although other investigators have implicated ERs in the neuroprotective effects of genistein (Zeng et al., 2004, Kajta et al., 2007), the ER subtype has not been determined. Experiments in primary cortical neurons by Linford and Dorsa, support a role for ERβ in genistein-dependent neuroprotection against thapsigargin-induced apoptosis (Linford and Dorsa, 2002). Sawada et al. (Sawada et al., 2000) also attributed the neuroprotective effects of genistein to ERβ in nigrostriatal dopamine neurons. However, it is clear that either ERα or ERβ can mediate E2-dependent neuroprotection in neuronal cell lines (Gollapudi and Oblinger, 1999, Fitzpatrick et al., 2002). Furthermore, some neuroprotective effects of estrogen-like compounds appear to be independent of their ability to bind ERs (Prokai and Simpkins, 2007). Although our results support a role for ERα, other mechanisms cannot be ruled out.

Among the mechanisms that contribute to estrogen-dependent neuroprotection is activation of intracellular kinase cascades including PI3K/AKT and MAPK. In vitro estrogen increases AKT phosphorylation and inhbits glutamate (Honda et al., 2000) and staurosporine-induced apoptosis in cortical neurons (Harms et al., 2001, Honda et al., 2001). In retinal neurons Akt also mediates estrogen-dependent neuroprotection from H2O2-induced apoptosis (Yu et al., 2004). AKT activation is also linked to estrogen-dependent neuroprotection in OGD exposed hippocampal organotypic cultures (Cimarosti et al., 2005) and chemical ischemia in cortical explants (Wilson et al., 2002).

In vivo PI3K inhibition prevent estrogen-dependent neuroprotection in 6-OHDA lesioned rats (Quesada et al., 2008), MPTP-treated mice (D'Astous et al., 2006), and global ischemia (Wang et al., 2006). However, PI3K/AKT also serves as an endogenous survival factor and inhibition of PI3K alone can increase cerebral ischemic injury (Noshita et al., 2001). Nevertheless, PI3K/AKT appear to play a role in neuroprotection by genistein and daidzein in response to OGD. This action in neurons is counter to the inhibition of AKT demonstrated with higher genistein doses in cancer cells (Sarkar and Li, 2002), and may represent the ability of estrogenic rather than tyrosine kinase inhibitor activity of these compounds.

In addition to a role for PI3K/AKT in phytoestrogen-dependent neuroprotection, inhibition of MAPK with PD98059 also prevented genistein inhibition of LDH release. High doses of genistein can induced apoptosis in primary cortical neurons, partially through MAPK activation, although only at doses greater than 10 μM (Linford et al., 2001). In contrast, lower does of genistein can inhibit glutamate-induced apoptosis in primary hippocampal neurons (Zhao et al., 2002) and thapsigargin-induced apoptosis in cortical neurons (Linford and Dorsa, 2002). However, these effects have not been shown to involve MAPK. In the present study, MAPK appears to be involved in genistein neuroprotection but not that seen with daidzein or equol.

Like estrogen, doses of isoflavones obtained by diet can be protective against a number of apoptotic challenges in vitro. The present study extends these observations to inhibition of several aspects of ischemic cell death, including glutamate, calcium, hypoxia, and OGD dependent cell death. At least part of this effect is mediated by an inhibition of caspase activity and depends on ER and PI3K, similar to estrogen-mediated neuroprotection. These results further support a role for soy isoflavones as neuroprotectants and possible alternatives to estrogen.

Acknowledgments

Research was supported by NIH R01AT001882 (D.A.S.). Portions of this data were presented previously at the 2006 Annual Meeting of the Society for Neuroscience.

Abbreviations

6-OHDA

6-hydroxydopamine

AKT

Akt serine-threonine kinase

B27

B-27 serum-free supplement

BCA

bicinchoninic acid

BSA

bovine serum albumin

C

celsius

Calcein AM

calcein acetoxymethylester

cm2

square centimeters

CO2

carbon dioxide

DAPI

4′,6-diamidino-2-phenylindole

DIV

days in vitro

DMSO

dimethyl sulfoxide

E2

17β-estradiol

E2-BSA

17β-estradiol conjugated to bovine serum albumin

ER

estrogen receptor

GAPDH

glyceraldehyde-3-phosphate dehydrogenase

GFAP

glial fibrillary acidic protein

GPR30

G-protein coupled receptor 30

H2O2

hydorgen peroxide

HBSS

Hank's buffered saline solution

KCN

potassium cyanide

kDa

kiloDalton

LDH

lactate dehydrogenase

MAP2

microtubule associated protein 2

MAPK

mitogen-activated protein kinase

MEK

mitogen-activated protein kinase kinase

MPTP

1-methyl 4-phenyl 1,2,3,6-tetrahydropyridine

mRNA

messenger RNA

NB

neurobasal medium

OGD

oxygen-glucose deprivation

PBS

phosphate buffered saline

PI3K

phosphoinositide 3-knase

RNA

ribonucleic acid

RT-PCR

reverse transcription polymerase chain reaction

SDS

sodium dodecyl sulfate

SERM

selective estrogen receptor modulator

Footnotes

No Supplementary Material

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Bibliography

  1. Aguado-Llera D, Arilla-Ferreiro E, Chowen JA, Argente J, Puebla-Jimenez L, Frago LM, Barrios V. 17beta-Estradiol protects depletion of rat temporal cortex somatostatinergic system by beta-amyloid. Neurobiol Aging. 2007;28:1396–1409. doi: 10.1016/j.neurobiolaging.2006.06.009. [DOI] [PubMed] [Google Scholar]
  2. Alonso de Lecinana M, Egido JA. Estrogens as neuroprotectants against ischemic stroke. Cerebrovasc Dis. 2006;21 2:48–53. doi: 10.1159/000091703. [DOI] [PubMed] [Google Scholar]
  3. Amantea D, Russo R, Bagetta G, Corasaniti MT. From clinical evidence to molecular mechanisms underlying neuroprotection afforded by estrogens. Pharmacol Res. 2005;52:119–132. doi: 10.1016/j.phrs.2005.03.002. [DOI] [PubMed] [Google Scholar]
  4. Bake S, Sohrabji F. 17beta-estradiol differentially regulates blood-brain barrier permeability in young and aging female rats. Endocrinology. 2004;145:5471–5475. doi: 10.1210/en.2004-0984. [DOI] [PubMed] [Google Scholar]
  5. Bramlett HM, Dietrich WD. Neuropathological protection after traumatic brain injury in intact female rats versus males or ovariectomized females. J Neurotrauma. 2001;18:891–900. doi: 10.1089/089771501750451811. [DOI] [PubMed] [Google Scholar]
  6. Brass LM. Hormone replacement therapy and stroke: clinical trials review. Stroke. 2004;35:2644–2647. doi: 10.1161/01.STR.0000143218.20061.ac. [DOI] [PubMed] [Google Scholar]
  7. Burguete MC, Torregrosa G, Perez-Asensio FJ, Castello-Ruiz M, Salom JB, Gil JV, Alborch E. Dietary phytoestrogens improve stroke outcome after transient focal cerebral ischemia in rats. Eur J Neurosci. 2006;23:703–710. doi: 10.1111/j.1460-9568.2006.04599.x. [DOI] [PubMed] [Google Scholar]
  8. Callier S, Morissette M, Grandbois M, Pelaprat D, Di Paolo T. Neuroprotective properties of 17beta-estradiol, progesterone, and raloxifene in MPTP C57Bl/6 mice. Synapse. 2001;41:131–138. doi: 10.1002/syn.1067. [DOI] [PubMed] [Google Scholar]
  9. Carswell HV, Macrae IM, Gallagher L, Harrop E, Horsburgh KJ. Neuroprotection by a selective estrogen receptor {beta} agonist in a mouse model of global ischemia. Am J Physiol Heart Circ Physiol. 2004;287:H1501–H1504. doi: 10.1152/ajpheart.00227.2004. [DOI] [PubMed] [Google Scholar]
  10. Chen S, Nilsen J, Brinton RD. Dose and temporal pattern of estrogen exposure determines neuroprotective outcome in hippocampal neurons: therapeutic implications. Endocrinology. 2006;147:5303–5313. doi: 10.1210/en.2006-0495. [DOI] [PubMed] [Google Scholar]
  11. Cimarosti H, Siqueira IR, Zamin LL, Nassif M, Balk R, Frozza R, Dalmaz C, Netto CA, Salbego C. Neuroprotection and protein damage prevention by estradiol replacement in rat hippocampal slices exposed to oxygen-glucose deprivation. Neurochem Res. 2005;30:583–589. doi: 10.1007/s11064-005-2693-1. [DOI] [PubMed] [Google Scholar]
  12. D'Astous M, Mendez P, Morissette M, Garcia-Segura LM, Di Paolo T. Implication of the phosphatidylinositol-3 kinase/protein kinase B signaling pathway in the neuroprotective effect of estradiol in the striatum of 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine mice. Mol Pharmacol. 2006;69:1492–1498. doi: 10.1124/mol.105.018671. [DOI] [PubMed] [Google Scholar]
  13. Dubal DB, Zhu H, Yu J, Rau SW, Shughrue PJ, Merchenthaler I, Kindy MS, Wise PM. Estrogen receptor α, not β, is a critical link in estradiol-mediated protection against brain injury. Proceedings of the National Academy of Sciences, USA; 2000. pp. 1952–1957. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Fitzpatrick JL, Mize AL, Wade CB, Harris JA, Shapiro RA, Dorsa DM. Estrogen-mediated neuroprotection against beta-amyloid toxicity requires expression of estrogen receptor alpha or beta and activation of the MAPK pathway. J Neurochem. 2002;82:674–682. doi: 10.1046/j.1471-4159.2002.01000.x. [DOI] [PubMed] [Google Scholar]
  15. Gibson CL, Gray LJ, Murphy SP, Bath PM. Estrogens and experimental ischemic stroke: a systematic review. J Cereb Blood Flow Metab. 2006;26:1103–1113. doi: 10.1038/sj.jcbfm.9600270. [DOI] [PubMed] [Google Scholar]
  16. Gollapudi L, Oblinger MM. Stable transfection of PC12 cells with estrogen receptor (ERα): protective effects of estrogen on cell survival after serum deprivation. Journal of neuroscience Research. 1999;56:99–108. doi: 10.1002/(SICI)1097-4547(19990401)56:1<99::AID-JNR13>3.0.CO;2-G. [DOI] [PubMed] [Google Scholar]
  17. Harms C, Lautenschlager M, Bergk A, Katchanov J, Freyer D, Kapinya K, Herwig U, Megow D, Dirnagl U, Weber JR, Hortnagl H. Differential mechanisms of neuroprotection by 17 beta-estradiol in apoptotic versus necrotic neurodegeneration. Journal of Neuroscience. 2001;21:2600–2609. doi: 10.1523/JNEUROSCI.21-08-02600.2001. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Heyer A, Hasselblatt M, von Ahsen N, Hafner H, Siren AL, Ehrenreich H. In vitro gender differences in neuronal survival on hypoxia and in 17beta-estradiol-mediated neuroprotection. J Cereb Blood Flow Metab. 2005;25:427–430. doi: 10.1038/sj.jcbfm.9600056. [DOI] [PubMed] [Google Scholar]
  19. Honda K, Sawada H, Kihara T, Urushitani M, Nakamizo T, Akaike A, Shimohama S. Phosphatidylinositol 3-kinase mediates neuroprotection by estrogen in cultured cortical neurons. J Neurosci Res. 2000;60:321–327. doi: 10.1002/(SICI)1097-4547(20000501)60:3<321::AID-JNR6>3.0.CO;2-T. [DOI] [PubMed] [Google Scholar]
  20. Honda K, Shimohama S, Sawada H, Kihara T, Nakamizo T, Shibasaki H, Akaike A. Nongenomic antiapoptotic signal transduction by estrogen in cultured cortical neurons. J Neurosci Res. 2001;64:466–475. doi: 10.1002/jnr.1098. [DOI] [PubMed] [Google Scholar]
  21. Hurn PD, Brass LM. Estrogen and stroke: a balanced analysis. Stroke. 2003;34:338–341. doi: 10.1161/01.str.0000054051.88378.25. [DOI] [PubMed] [Google Scholar]
  22. Jones PA, May GR, McLuckie JA, Iwashita A, Sharkey J. Apoptosis is not an invariable component of in vitro models of cortical cerebral ischaemia. Cell Res. 2004;14:241–250. doi: 10.1038/sj.cr.7290225. [DOI] [PubMed] [Google Scholar]
  23. Jover T, Tanaka H, Calderone A, Oguro K, Bennett MV, Etgen AM, Zukin RS. Estrogen protects against global ischemia-induced neuronal death and prevents activation of apoptotic signaling cascades in the hippocampal CA1. J Neurosci. 2002;22:2115–2124. doi: 10.1523/JNEUROSCI.22-06-02115.2002. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Kajta M, Domin H, Grynkiewicz G, Lason W. Genistein inhibits glutamate-induced apoptotic processes in primary neuronal cell cultures: an involvement of aryl hydrocarbon receptor and estrogen receptor/glycogen synthase kinase-3beta intracellular signaling pathway. Neuroscience. 2007;145:592–604. doi: 10.1016/j.neuroscience.2006.11.059. [DOI] [PubMed] [Google Scholar]
  25. Krause DN, Duckles SP, Pelligrino DA. Influence of sex steroid hormones on cerebrovascular function. J Appl Physiol. 2006;101:1252–1261. doi: 10.1152/japplphysiol.01095.2005. [DOI] [PubMed] [Google Scholar]
  26. Linford NJ, Dorsa DM. 17beta-Estradiol and the phytoestrogen genistein attenuate neuronal apoptosis induced by the endoplasmic reticulum calcium-ATPase inhibitor thapsigargin. Steroids. 2002;67:1029–1040. doi: 10.1016/s0039-128x(02)00062-4. [DOI] [PubMed] [Google Scholar]
  27. Linford NJ, Yang Y, Cook DG, Dorsa DM. Neuronal apoptosis resulting from high doses of the isoflavone genistein: role for calcium and p42/44 mitogen-activated protein kinase. J Pharmacol Exp Ther. 2001;299:67–75. [PubMed] [Google Scholar]
  28. Mahesh VB, Dhandapani KM, Brann DW. Role of astrocytes in reproduction and neuroprotection. Mol Cell Endocrinol. 2006;246:1–9. doi: 10.1016/j.mce.2005.11.017. [DOI] [PubMed] [Google Scholar]
  29. McCullough LD, Hurn PD. Estrogen and ischemic neuroprotection: an integrated view. Trends Endocrinol Metab. 2003;14:228–235. doi: 10.1016/s1043-2760(03)00076-6. [DOI] [PubMed] [Google Scholar]
  30. Mendelowitsch A, Ritz MF, Ros J, Langemann H, Gratzl O. 17beta-Estradiol reduces cortical lesion size in the glutamate excitotoxicity model by enhancing extracellular lactate: a new neuroprotective pathway. Brain Res. 2001;901:230–236. doi: 10.1016/s0006-8993(01)02359-9. [DOI] [PubMed] [Google Scholar]
  31. Mize AL, Shapiro RA, Dorsa DM. Estrogen receptor-mediated neuroprotection from oxidative stress requires activation of the mitogen-activated protein kinase pathway. Endocrinology. 2003;144:306–312. doi: 10.1210/en.2002-220698. [DOI] [PubMed] [Google Scholar]
  32. Morissette M, Jourdain S, Al Sweidi S, Menniti FS, Ramirez AD, Di Paolo T. Role of estrogen receptors in neuroprotection by estradiol against MPTP toxicity. Neuropharmacology. 2007;52:1509–1520. doi: 10.1016/j.neuropharm.2007.02.004. [DOI] [PubMed] [Google Scholar]
  33. Nilsen J, Brinton RD. Impact of progestins on estrogen-induced neuroprotection: synergy by progesterone and 19-norprogesterone and antagonism by medroxyprogesterone acetate. Endocrinology. 2002;143:205–212. doi: 10.1210/endo.143.1.8582. [DOI] [PubMed] [Google Scholar]
  34. Noshita N, Lewen A, Sugawara T, Chan PH. Evidence of phosphorylation of Akt and neuronal survival after transient focal cerebral ischemia in mice. J Cereb Blood Flow Metab. 2001;21:1442–1450. doi: 10.1097/00004647-200112000-00009. [DOI] [PubMed] [Google Scholar]
  35. O'Donnell ME, Lam TI, Tran LQ, Foroutan S, Anderson SE. Estradiol reduces activity of the blood-brain barrier Na-K-Cl cotransporter and decreases edema formation in permanent middle cerebral artery occlusion. J Cereb Blood Flow Metab. 2006;26:1234–1249. doi: 10.1038/sj.jcbfm.9600278. [DOI] [PubMed] [Google Scholar]
  36. Pelligrino DA, Santizo R, Baughman VL, Wang Q. Cerebral vasodilating capacity during forebrain ischemia: effects of chronic estrogen depletion and repletion and the role of neuronal nitric oxide synthase. Neuroreport. 1998;9:3285–3291. doi: 10.1097/00001756-199810050-00026. [DOI] [PubMed] [Google Scholar]
  37. Prokai L, Simpkins JW. Structure-nongenomic neuroprotection relationship of estrogens and estrogen-derived compounds. Pharmacol Ther. 2007;114:1–12. doi: 10.1016/j.pharmthera.2007.01.006. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Quesada A, Lee BY, Micevych PE. PI3 kinase/Akt activation mediates estrogen and IGF-1 nigral DA neuronal neuroprotection against a unilateral rat model of Parkinson's disease. Dev Neurobiol. 2008;68:632–644. doi: 10.1002/dneu.20609. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Sampei K, Goto S, Alkayed NJ, Crain BJ, Korach KS, Traystman RJ, Demas GE, Nelson RJ, Hurn PD. Stroke in estrogen receptor-alpha-deficient mice. Stroke. 2000;31:738–743. doi: 10.1161/01.str.31.3.738. discussion 744. [DOI] [PubMed] [Google Scholar]
  40. Sarkar FH, Li Y. Mechanisms of cancer chemoprevention by soy isoflavone genistein. Cancer Metastasis Rev. 2002;21:265–280. doi: 10.1023/a:1021210910821. [DOI] [PubMed] [Google Scholar]
  41. Sawada H, Ibi M, Kihara T, Urushitani M, Honda K, Nakanishi M, Akaike A, Shimohama S. Mechanisms of antiapoptotic effects of estrogens in nigral dopaminergic neurons. FASEB Journal. 2000;14:1202–1214. doi: 10.1096/fasebj.14.9.1202. [DOI] [PubMed] [Google Scholar]
  42. Schreihofer DA. Transcriptional regulation by phytoestrogens in neuronal cell lines. Mol Cell Endocrinol. 2005;231:13–22. doi: 10.1016/j.mce.2004.12.006. [DOI] [PubMed] [Google Scholar]
  43. Schreihofer DA, Do KD, Schreihofer AM. High-soy diet decreases infarct size after permanent middle cerebral artery occlusion in female rats. Am J Physiol Regul Integr Comp Physiol. 2005;289:R103–108. doi: 10.1152/ajpregu.00642.2004. [DOI] [PubMed] [Google Scholar]
  44. Setchell KD. Soy isoflavones--benefits and risks from nature's selective estrogen receptor modulators (SERMs) J Am Coll Nutr. 2001;20:354S–362S. doi: 10.1080/07315724.2001.10719168. discussion 381S-383S. [DOI] [PubMed] [Google Scholar]
  45. Singer CA, Figueroa-Masot XA, Batchelor RH, Dorsa DM. The mitogen-activated protein kinase pathway mediates estrogen neuroprotection after glutamate toxicity in primary cortical neurons. Journal of Neuroscience. 1999;19:2455–2463. doi: 10.1523/JNEUROSCI.19-07-02455.1999. [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Sribnick EA, Ray SK, Nowak MW, Li L, Banik NL. 17beta-estradiol attenuates glutamate-induced apoptosis and preserves electrophysiologic function in primary cortical neurons. J Neurosci Res. 2004;76:688–696. doi: 10.1002/jnr.20124. [DOI] [PubMed] [Google Scholar]
  47. Suzuki S, Brown CM, Dela Cruz CD, Yang E, Bridwell DA, Wise PM. Timing of estrogen therapy after ovariectomy dictates the efficacy of its neuroprotective and antiinflammatory actions. Proc Natl Acad Sci U S A. 2007;104:6013–6018. doi: 10.1073/pnas.0610394104. [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. Trieu VN, Uckun FM. Genistein is neuroprotective in murine models of familial amyotrophic lateral sclerosis and stroke. Biochem Biophys Res Commun. 1999;258:685–688. doi: 10.1006/bbrc.1999.0577. [DOI] [PubMed] [Google Scholar]
  49. Wang R, Zhang QG, Han D, Xu J, Lu Q, Zhang GY. Inhibition of MLK3-MKK4/7-JNK1/2 pathway by Akt1 in exogenous estrogen-induced neuroprotection against transient global cerebral ischemia by a non-genomic mechanism in male rats. J Neurochem. 2006;99:1543–1554. doi: 10.1111/j.1471-4159.2006.04201.x. [DOI] [PubMed] [Google Scholar]
  50. Wassertheil-Smoller S, Hendrix SL, Limacher M, Heiss G, Kooperberg C, Baird A, Kotchen T, Curb JD, Black H, Rossouw JE, Aragaki A, Safford M, Stein E, Laowattana S, Mysiw WJ. Effect of estrogen plus progestin on stroke in postmenopausal women: the Women's Health Initiative: a randomized trial. Jama. 2003;289:2673–2684. doi: 10.1001/jama.289.20.2673. [DOI] [PubMed] [Google Scholar]
  51. Wathen CN. Alternatives to hormone replacement therapy: a multi-method study of women's experiences. Complement Ther Med. 2006;14:185–192. doi: 10.1016/j.ctim.2005.11.003. [DOI] [PubMed] [Google Scholar]
  52. Wilson ME, Dubal DB, Wise PM. Estradiol protects against injury-induced cell death in cortical explant cultures: a role for estrogen receptors. Brain Research. 2000;873:235–242. doi: 10.1016/s0006-8993(00)02479-3. [DOI] [PubMed] [Google Scholar]
  53. Wilson ME, Liu Y, Wise PM. Estradiol enhances Akt activation in cortical explant cultures following neuronal injury. Brain Res Mol Brain Res. 2002;102:48–54. doi: 10.1016/s0169-328x(02)00181-x. [DOI] [PubMed] [Google Scholar]
  54. Yu X, Rajala RV, McGinnis JF, Li F, Anderson RE, Yan X, Li S, Elias RV, Knapp RR, Zhou X, Cao W. Involvement of insulin/phosphoinositide 3-kinase/Akt signal pathway in 17 beta-estradiol-mediated neuroprotection. J Biol Chem. 2004;279:13086–13094. doi: 10.1074/jbc.M313283200. [DOI] [PubMed] [Google Scholar]
  55. Zeng H, Chen Q, Zhao B. Genistein ameliorates beta-amyloid peptide (25-35)-induced hippocampal neuronal apoptosis. Free Radic Biol Med. 2004;36:180–188. doi: 10.1016/j.freeradbiomed.2003.10.018. [DOI] [PubMed] [Google Scholar]
  56. Zhao L, Chen Q, Diaz Brinton R. Neuroprotective and neurotrophic efficacy of phytoestrogens in cultured hippocampal neurons. Exp Biol Med (Maywood) 2002;227:509–519. doi: 10.1177/153537020222700716. [DOI] [PubMed] [Google Scholar]

RESOURCES