Cerebroprotective Effect of 17β-Estradiol Replacement Therapy in Ovariectomy-Induced Post-Menopausal Rats Subjected to Ischemic Stroke: Role of MAPK/ERK1/2 Pathway and PI3K-Independent Akt Activation

Abstract

Despite the overwhelming advances in the understanding of the pathogenesis of stroke, a devastating disease affecting millions of people worldwide, currently there are only a limited number of effective treatments available. Preclinical and clinical studies show that stroke is a sexually dimorphic disorder, affecting males and females differently. Strong experimental evidence indicates that estrogen may play a role in this difference and that exogenous 17β-estradiol (E2) is neuroprotective against stroke in both male and female rodents. However, the molecular mechanisms by which E2 intervenes in ischemia-induced cell death, revealing these sex differences, remain unclear. The present study was aimed to determine, in female rats, the molecular mechanisms of two well-known pro-survival signaling pathways, MAPK/ERK1/2 and PI3K/Akt, that mediate E2 neuroprotection in response to acute ischemic stroke. E2 pretreatment reduced brain damage and attenuated apoptotic cell death in ovariectomized female rats after an ischemic insult. Moreover, E2 decreased phosphorylation of ERK1/2 and prevented ischemia/reperfusion-induced dephosphorylation of both Akt and the pro-apoptotic protein, BAD. However, MAPK/ERK1/2 inhibitor PD98059, but not the PI3K inhibitor LY294002, attenuated E2 neuroprotection. Thus, these results suggested that E2 pretreatment in ovariectomized female rats modulates MAPK/ERK1/2 and activates Akt independently of PI3K to promote cerebroprotection in ischemic stroke. A better understanding of the mechanisms and the influence of E2 in the female sex paves the way for the design of future successful hormone replacement therapies.

Keywords: 17β-estradiol; MAPK/ERK; PI3K/Akt; apoptosis; ischemic stroke; neuronal death; neuroprotection.

Figure 1

Estradiol decreases infarct volume following tMCAO. (A) Effect of E2-treatment on ischemic stroke-induced neurofunctional impairment in ovariectomized female rats subjected to tMCAO (p = 0.27, U = 43, Mann–Whitney test). (B) Effect of E2-treatment on serum E2 levels. Significantly different from placebo-treated group (*** p < 0.001, t = 6.9, DF = 15, Student’s t-test). (C) Representative images of the TTC-stained third coronal brain slice (0.2/-1.8 mm from Bregma) from placebo- and E2-treated animals, and summary data showing total infarct volumes (expressed as a percentage of the corresponding non-ischemic hemisphere). Significantly different from placebo-treated group (** p < 0.01, t = 4, DF = 15, Student’s t-test). (D) Summary data showing cortical and subcortical infarct volumes. Two-way ANOVA: treatment (E2 vs. placebo, p < 0.001, F_1,30 = 38), brain region (subcortical vs. cortical, p < 0.001, F_1,30 = 28), and interaction (p < 0.05, F_1,30 = 7.1). Post hoc Sidak’s multiple comparisons test: significantly different from cortical infarct volume of placebo-treated animals (*** p < 0.001) and significantly different from subcortical infarct volume of placebo-treated animals (^# p < 0.05). Data are median (Q1, Q3) (box-and-whisker plot) or mean ± SEM (bar graphs) of individual data points. E2, 17β-estradiol. TTC, 2,3,5-triphenyltetrazolium chloride. tMCAO, transient middle cerebral artery occlusion.

Figure 2

Estradiol attenuates ischemia/reperfusion-induced TUNEL-detected apoptosis. TUNEL detection of apoptotic DNA cleavage in brains from placebo- and E2-treated ovariectomized female rats subjected to tMCAO. (A) Lack of TUNEL-labeling (a,e) in a cortex field of the non-ischemic hemisphere (○), and double-labeling (TUNEL, (b,f), and TUNEL + DAPI merged images, (c,g)) in a cortex field of the ischemic hemisphere (●) of representative brain sections. Scale bar, 50 µm. Insets of TUNEL + DAPI merged images (d,h) are shown at higher magnification. Scale bar, 25 µm. (B) TUNEL-positive cell counts expressed as a percentage of the corresponding DAPI-stained nuclei in three selected fields of the ischemic hemisphere (●). Significantly different from placebo group (* p < 0.05, t = 3.1, DF = 5, Student’s t-test). Data are mean ± SEM of individual data points. E2, 17β-estradiol. tMCAO, transient middle cerebral artery occlusion. TUNEL, terminal deoxynucleotidyl transferase dUTP nick end labeling.

Figure 3

Estradiol inhibits ischemia/reperfusion-induced caspase-3-mediated apoptosis. Apoptosis executor caspase-3 expression in brains from placebo- and E2-treated ovariectomized female rats subjected to tMCAO. (A) FAM-DEVD-FMK labeling in a cortex field of the non-ischemic hemisphere (○, (a,c)) and the ischemic hemisphere (●, (b,d)) of representative brain sections. Scale bar, 25 µm. (B) Fluorescence quantification within three fields of the ischemic cortex (●), expressed as a percentage of the field area. Significantly different from placebo group (** p < 0.01, t = 4.8, DF = 6, Student’s t-test). Data are mean ± SEM of individual data points. (C) Representative Western blots and (D) relative abundance of activated caspase-3 p17 form in whole-cell lysates of non-ischemic (NI) and ischemic (I) hemispheres. Two-way ANOVA: ischemic vs. non-ischemic hemisphere p < 0.01, F_1,28 = 9.7. Post hoc Sidak’s multiple comparisons test: significantly different from non-ischemic placebo group (** p < 0.01) or from ischemic placebo group (^# p < 0.05). Data are mean ± SEM of individual data points (normalized to β-actin). E2, 17β-estradiol. tMCAO, transient middle cerebral artery occlusion.

Figure 4

Estradiol decreases phosphorylation of ERK1/2. Phosphorylated extracellular signal-regulated kinase 1/2 (pERK1/2) protein expression in brains from placebo- and E2-treated ovariectomized female rats subjected to tMCAO. (A) Representative Western blots and (B) relative abundance of pERK1 and (C) pERK2 in whole-cell lysates of non-ischemic (NI) and ischemic (I) hemispheres. Two-way ANOVA for pERK1: brain hemisphere (ischemic vs. non-ischemic, p < 0.001, F_1,30 = 18.1), treatment (E2 vs. placebo, p < 0.001, F_1,30 = 84.6), and interaction (p < 0.001, F_1,30 = 15.5). Two-way ANOVA for pERK2: brain hemisphere (ischemic vs. non-ischemic, p < 0.05, F_1,30 = 4.7) and treatment (E2 vs. placebo, p < 0.001, F_1,30 = 28.1). Post hoc Sidak’s multiple comparisons tests: significantly different from non-ischemic placebo group (* p < 0.05, ** p < 0.01, and *** p < 0.001), or from ischemic placebo group (^### p < 0.001). Data are mean ± SEM of individual data points (normalized to total ERK1/2). E2, 17β-estradiol. tMCAO, transient middle cerebral artery occlusion.

Figure 5

Estradiol blunts the ischemia/reperfusion-induced decrease in Akt phosphorylation. Phosphorylated serine-threonine protein kinase B (pAkt) protein expression in brains from placebo- and E2-treated ovariectomized female rats subjected to tMCAO. (A) Representative Western blots, and (B) relative abundance of pAkt in whole-cell lysates of non-ischemic (NI) and ischemic (I) hemispheres. Two-way ANOVA: brain hemisphere (ischemic vs. non-ischemic, p < 0.001, F_1,22 = 15) and treatment (E2 vs. placebo, p < 0.001, F_1,22 = 26). Post hoc Sidak’s multiple comparisons test: significantly different from non-ischemic placebo group (* p < 0.05 and ** p < 0.01), or from ischemic placebo group (^### p < 0.001). Data are mean ± SEM of individual data points (normalized to total Akt). E2, 17β-estradiol. tMCAO, transient middle cerebral artery occlusion.

Figure 6

Estradiol increases the CTMP protein expression. Carboxy-terminal modulator protein (CTMP) protein expression in brains from placebo- and E2-treated ovariectomized female rats subjected to tMCAO. (A) Representative Western blots and (B) relative abundance of CTMP in whole-cell lysates of non-ischemic (NI) and ischemic (I) hemispheres. Two-way ANOVA: E2 vs. placebo, p = 0.08, F_1,24 = 3.1). Post hoc Sidak’s multiple comparisons test: significantly different from non-ischemic placebo group (* p < 0.05). Data are mean ± SEM of individual data points (normalized to β-actin). E2, 17β-estradiol. tMCAO, transient middle cerebral artery occlusion.

Figure 7

Estradiol increases phosphorylation of the pro-apoptotic protein BAD. Phosphorylated BAD (pBAD) protein expression in brains from placebo- and E2-treated ovariectomized female rats subjected to tMCAO. (A) Representative Western blots and (B) relative abundance of pBAD in whole-cell lysates of non-ischemic (NI) and ischemic (I) hemispheres. Two-way ANOVA: brain hemisphere (ischemic vs. non-ischemic, p < 0.01, F_1,18 = 15) and treatment (E2 vs. placebo, p < 0.001, F_1,18 = 22). Post hoc Sidak’s multiple comparisons test: significantly different from non-ischemic placebo group (** p < 0.01), from ischemic placebo-treated group (^# p < 0.05), or from non-ischemic E2 group (^& p < 0.05). Data are mean ± SEM of individual data points (normalized to total BAD). E2, 17β-estradiol. tMCAO, transient middle cerebral artery occlusion.

Figure 8

The MEK inhibitor PD98059, but not the PI3K inhibitor LY294002, attenuates estradiol protection. Infarct sizes in brains from placebo- and E2-treated ovariectomized female rats subjected to tMCAO. (A) Representative images of the TTC-stained third coronal brain slice from placebo- (−) and E2-treated (+) animals, treated with vehicle, PD98059 or LY294002. (B) Summary data showing total infarct volumes, expressed as a percentage of the corresponding non-ischemic hemisphere. Two-way ANOVA for estrogen treatment (E2 vs. placebo, p < 0.0001, F_1,31 = 32.5), and inhibitors treatment (PD98059 and LY294002 vs. vehicle, p < 0.01, F_2,31 = 7.9). Post hoc Sidak’s multiple comparisons tests: significantly different from placebo group (* p < 0.05 and ** p < 0.01) or from vehicle-treated group (^# p < 0.05). Data are mean ± SEM of individual data points. E2, 17β-estradiol. tMCAO, transient middle cerebral artery occlusion.