Estrogen Regulates Angiotensin II Receptor Expression Patterns and Protects the Heart from Ischemic Injury in Female Rats

Abstract

Previous studies have shown that female offspring are resistant to fetal stress-induced programming of ischemic-sensitive phenotype in the heart; however, the mechanisms responsible remain unclear. The present study tested the hypothesis that estrogen plays a role in protecting females in fetal programming of increased heart vulnerability. Pregnant rats were divided into normoxic and hypoxic (10.5% O2 from Day 15 to 21 of gestation) groups. Ovariectomy (OVX) and estrogen (E2) replacement were performed in 8-wk-old female offspring. Hearts of 4-mo-old females were subjected to ischemia and reperfusion injury in a Langendorff preparation. OVX significantly decreased postischemic recovery of left ventricular function and increased myocardial infarction, and no difference was observed between normoxic and hypoxic groups. The effect of OVX was rescued by E2 replacement. OVX decreased the binding of glucocorticoid receptor (GR) to glucocorticoid response elements at angiotensin II type 1 (Agtr1) and type 2 (Agtr2) receptor promoters, resulting in a decrease in Agtr1 and an increase in Agtr2 in the heart. Additionally, OVX decreased estrogen receptor (ER) expression in the heart and inhibited ER/GR interaction in binding to glucocorticoid response elements at the promoters. Consistent with the changes in Agtrs, OVX significantly decreased Prkce abundance in the heart. These OVX-induced changes were abrogated by E2 replacement. The results indicate that estrogen is not directly responsible for the sex dimorphism in fetal programming of heart ischemic vulnerability but suggest a novel mechanism of estrogen in regulating cardiac Agtr1/Agtr2 expression patterns and protecting female hearts against ischemia and reperfusion injury.

Keywords: angiotensin II receptor; estrogen; heart; ischemic injury.

FIG. 1

Effect of ovariectomy (OVX) and estrogen replacement (OVX-E₂) on postischemic recovery of LV function. Isolated hearts were studied in a Langendorff preparation in 4-mo-old females of normoxic control and antenatal hypoxic animals. Data are means ± SEM (n = 5). LV functional recovery data (A–C) were expressed as fold change of preischemic baseline values. Infarct size data (E) was expressed as a percentage of the total LV weight. Data of LDH release (F) were expressed as the area under curve (AUC) in coronary effluent collected from 30 sec before the onset of ischemia to 30 min of reperfusion. Data in A–D were analyzed by repeated measures two-way ANOVA with treatment as one factor and time as another, followed by Bonferroni posttests. Data in E and F were analyzed by two-way ANOVA with treatment as one factor and hypoxia as another, followed by Bonferroni posttests. *P < 0.05, OVX versus sham control or E₂ replacement in both normoxic and hypoxic animals.

FIG. 2

Effect of ovariectomy (OVX) and estrogen replacement (OVX-E₂) on Agtr1 and Agtr2 protein and mRNA abundance in normoxic females. LVs were isolated from 4-mo-old normoxic females. A) Agtr1 and Agtr2 protein abundance was determined by Western blot analysis (n = 4). B) Agtr1a, Agtr1b, and Agtr2 mRNA abundance was determined by real-time RT-PCR (n = 5). Data are means ± SEM. Data were analyzed by one-way ANOVA followed by Neuman-Keuls posttests, within each receptor subtype. *P < 0.05, OVX versus sham control or E₂ replacement.

FIG. 3

Effect of ovariectomy (OVX) and estrogen replacement (OVX-E₂) on Prkce and Prkcd protein and mRNA abundance in normoxic females. LVs were isolated from 4-mo-old normoxic females. A) Protein abundance of Prkce and Prkcd, was determined by Western blot analysis (n = 4). B) Prkce and Prkcd mRNA abundance was determined by real-time RT-PCR (n = 5). Data are means ± SEM. Data were analyzed by one-way ANOVA followed by Neuman-Keuls posttests, within each isozyme. *P < 0.05, OVX versus sham control or E₂ replacement.

FIG. 4

Effect of ovariectomy (OVX) and estrogen replacement (OVX-E₂) on GR-binding affinity to GREs at Agtr1a and Agtr2 promoters in normoxic females. LVs were isolated from 4-mo-old normoxic females. The GR-binding affinity to GRE2 at Agtr1a promoter (A) and GRE4 at Agtr2 promoter (B) was determined in a competition EMSA performed in pooled nuclear extracts from LVs (n = 5 in each group) with increasing ratios of unlabeled to labeled oligonucleotides encompassing GRE2 at Agtr1a promoter or GRE4 at Agtr2 promoter.

FIG. 5

Effect of ovariectomy (OVX) and estrogen replacement (OVX-E₂) on GR binding to GREs at Agtr1a and Agtr2 promoters in normoxic females. LVs were isolated from 4-mo-old normoxic females. GR binding to GREs at Agtr1a promoter (A) and GREs at Agtr2 promoter (B) was determined with ChIP assay using a GR antibody. Data are means ± SEM (n = 5 in each group). Data were analyzed by one-way ANOVA followed by Neuman-Keuls posttests, within each GRE site. *P < 0.05, OVX versus sham control or E₂ replacement.

FIG. 6

Effect of ovariectomy (OVX) and estrogen replacement (OVX-E₂) on ERα and ERβ protein abundance in normoxic females. LVs were isolated from 4-mo-old normoxic females. Protein abundance of ERα and ERβ was determined by Western blot analysis. Data are means ± SEM (n = 4). Data were analyzed by one-way ANOVA followed by Neuman-Keuls posttests, within each receptor subtype. *P < 0.05, OVX versus sham control or E₂ replacement.

FIG. 7

Effect of ovariectomy (OVX) and estrogen replacement (OVX-E₂) on ER/GR cross talk in binding to GREs at Agtr1a and Agtr2 promoters in normoxic females. LVs were isolated from 4-mo-old normoxic females. Cross talk binding of ER and GR to GREs at Agtr1a and Agtr2 promoters was determined by Re-ChIP assay with first GR pull-down and second ERα pull-down in pooled samples of LVs (n = 5 in each group). The blots were run on one membrane for each GRE. To be consistent with other figures, the presentation of the blot was reordered.