Progesterone inhibits estrogen-mediated neuroprotection against excitotoxicity by down-regulating estrogen receptor-β

Abstract

While both 17β-estradiol (E2) and progesterone (P4) are neuroprotective in several experimental paradigms, P4 also counteracts E2 neuroprotective effects. We recently reported that a 4-h treatment of cultured hippocampal slices with P4 following a prolonged (20 h) treatment with E2 eliminated estrogenic neuroprotection against NMDA toxicity and induction of brain-derived neurotrophic factor (BDNF) expression. In the present study, we evaluated the effects of the same treatment on levels of estrogen receptors, ERα and ERβ, and BDNF using a similar paradigm. E2 treatment resulted in elevated ERβ mRNA and protein levels, did not modify ERα mRNA, but increased ERα protein levels, and increased BDNF mRNA levels. P4 reversed E2-elicited increases in ERβ mRNA and protein levels, in ERα protein levels, and in BDNF mRNA levels. Experiments with an ERβ-specific antagonist, PHTPP, and specific agonists of ERα and ERβ, propylpyrazoletriol and diarylpropionitrile, respectively, indicated that E2-mediated neuroprotection against NMDA toxicity was, at least in part, mediated via ERβ receptor. In support of this conclusion, E2 did not protect against NMDA toxicity in cultured hippocampal slices from ERβ-/- mice. Thus, E2-mediated neuroprotection against NMDA toxicity may be because of estrogenic induction of BDNF via its ERβ receptor, and P4-mediated inhibition of E2 neuroprotective effects treatment to P4-induced down-regulation of ERβ and BDNF.

Figure 1. Effect of E2 and an ERβ antagonist PHTPP on NMDA-induced neurotoxicity in cultured hippocampal slices

Lactate dehydrogenase (LDH) release in the medium measured 24 h after NMDA (N) treatment. Medium was collected 24 h after initiation of NMDA treatment, and LDH activity was determined as described under Material and Methods. Results are means ± S.E.M of at least 3 experiments and are expressed as percent of release in NMDA-treated slices. Addition of the 1 μM PHTPP results in partial reversal of E2-mediated neuroprotection (E2+PHTPP+N) compared to E2-treated slices (E2+N). Statistical significance was analyzed by ANOVA followed by Tukey’s test for individual comparisons. *p < 0.01 vs. NMDA; †p= n.s. vs. NMDA.

Figure 2. ERβ but not ERα activation mediates E2 neuroprotection against NMDA toxicity in cultured hippocampal slices

Images of slices treated with selective ER agonists were analyzed as described under Materials and Methods. Slices were treated with 10nm and 1nm DPN or 100nM and 10nM PPT with and without NMDA (N).Regions analyzed (CA1, CA3 and DG) are described previously (Aguirre and Baudry, 2009). Results are expressed in percent of fluorescent intensity measured in slices treated with NMDA (50 μM) for 24 h, and are means ± S.E.M. of 12–14 images obtained in 3–5 independent experiments. Only DPN treatment at both concentrations reversed NMDA-mediated PI uptake increase. Significant results are seen in CA1 and DG subfields. Statistical significance was analyzed by ANOVA followed by Tukey’s test for individual comparisons. ** p < 0.01 as compared to NMDA treatment alone. Cont = vehicle, DMSO; N= 50 μM NMDA as described in methods section.

Figure 3. Effects of E2 and P4 on ERα and ERβ mRNA levels in cultured hippocampal slices

Cultured hippocampal slices were treated with E2 (10 nM), P4 (10 nM) or their combination according to the protocol described in Materials and Methods. Control slices were treated with vehicle (DMSO). Slices were collected at the end of treatment, and processed for determination of ERβ and Erα mRNA levels with RT-PCR. (a). Representative agarose gels for RT-PCR products showing mRNA levels of ERα (top panel), ERβ (middle panel) and β-actin (lower panel) for different treatment conditions. Quantitative analysis of PCR data showing (b) ERα mRNA levels and (c) ERβ levels, for different treatment conditions (quantification of RT-PCR was performed as described under Materials and Methods). Statistical significance was analyzed by ANOVA followed by Tukey’s test for individual comparisons. ** p < 0.01 as compared to vehicle-treated samples (Cont); † p<0.001 as compared to E2-treated values.

Figure 4. Effects of E2 and P4 on ERα and ERβ protein levels in cultured hippocampal slices

Cultured hippocampal slices were treated with E2 (10 nM), P4 (10 nM) or E2 + P4 (EP) as described in Figure 3. At the end of treatment, slices were collected and processed for western blot analysis of ERα (a) and ERβ (b). Levels of ERα and ERβ were corrected with those of β-actin and results were expressed as percent of values found in vehicle-treated slices; results are means ± S.E.M. of 4–5 experiments. Statistical significance was analyzed by ANOVA followed by Tukey’s test for individual comparisons. *** p < 0.001 as compared to vehicle-treated slices; * p < 0.05 as compared to vehicle-treated slices; †p<0.05 as compared to E2-treated values. (c) Immunocytochemistry results illustrate that E2 increased fluorescence of ERβ in slices treated with E2 alone, whereas concomitant treatment with E2 + P4 reversed this effect (A: Veh; B: E2; C: P4; D: E2 + P4). (Zooming in clearly indicates that immunoreactivity is real and not background).

Figure 5. Effects of P4 treatment on DPN-mediated neuroprotection against NMDA neurotoxicity in cultured hippocampal slices

Images of slices treated with selective ER agonists were analyzed as described under Materials and Methods. Results of regions analyzed (CA1 (a), CA3 (b) and DG (c)) are expressed in percent of fluorescent intensity measured in slices treated with 50 μM NMDA for 24 h and are means ± S.E.M of 12–14 images obtained in 3–5 independent experiments. Addition of P4 significantly reversed DPN-mediated neuroprotection against NMDA toxicity in CA1 and CA3, with a trend to do the same in DG. * p <0.01, as compared to NMDA; †p <0.01 as compared to 10 nM DPN + N.

Figure 6. Effect of E2 and P4 treatment on BDNF mRNA levels and effects of DPN, PPT, and P4 treatment on BDNF protein levels in cultured hippocampal slices

Cultured hippocampal slices were treated with E2 (10 nM), P4 (10 nM) or their combination according to the protocol previously described in Figure 3. Control slices were treated with vehicle (DMSO). Slices were collected at the end of treatment, and processed for determination of BDNF mRNA with RT-PCR. (a) Representative northern blots. (b) Results were normalized to levels of β-actin and were expressed as percent of control; they are means ± S.E.M. of at least 3 experiments. Statistical significance was analyzed by ANOVA followed by Tukey’s test for individual comparisons. ** p < 0.01 as compared to control to vehicle-treated values (Cont); †p < 0.01 as compared to E2-treated values. (c) Cultured hippocampal slices were treated with 100 nM PPT or 10 nM DPN in the presence or absence of 10 nM P4. At the end of treatment, slices were collected and processed for western blot analysis of BDNF and β-Actin. Levels of BDNF were corrected with β-Actin and expressed as percent of values found in vehicle (Cont)- treated slices. Results are means ±S.E.M. of 4–5 experiments. Statistical significance was analyzed by ANOVA followed by Tukey’s test for individual comparisons. *p< 0.01 as compared to vehicle-treated values; †p <0.01 as compared with 10 nM DPN values.

Figure 7. E2 treatment of cultured hippocampal slices from wild-type but not ERβ-KO mice results in neuroprotection against NMDA toxicity

Images of slices treated with E2 were analyzed as described under Materials and Methods. (a) LDH release in medium. (b) PI uptake in cultured mouse hippocampal slices. Results are expressed as percent of values found in control slices from cultured hippocampal slices from wild-type (WT) and ERβ-ko (ERβ−/−) mice, and are means ± S.E.M. of 6 independent experiments. * p <0.001, as compared to control values; †p <0.001 as compared to NMDA-treated values.

Figure 8. Proposed mechanism of P4-E2 interaction

P4 may be exerting its antagonistic effects on estrogenic neuroprotection and BDNF induction by destabilizing or degrading ERβ mRNA (dashed line). In this proposed mechanism, E2 is neuroprotective via activation of ERβ receptors, and P4 downregulates ERβ resulting in reversal of neuroprotection against NMDA.