Aldosterone's rapid, nongenomic effects are mediated by striatin: a modulator of aldosterone's effect on estrogen action

Abstract

The cellular responses to steroids are mediated by 2 general mechanisms: genomic and rapid/nongenomic effects. Identification of the mechanisms underlying aldosterone (ALDO)'s rapid vs their genomic actions is difficult to study, and these mechanisms are not clearly understood. Recent data suggest that striatin is a mediator of nongenomic effects of estrogen. We explored the hypothesis that striatin is an intermediary of the rapid/nongenomic effects of ALDO and that striatin serves as a novel link between the actions of the mineralocorticoid and estrogen receptors. In human and mouse endothelial cells, ALDO promoted an increase in phosphorylated extracellular signal-regulated protein kinases 1/2 (pERK) that peaked at 15 minutes. In addition, we found that striatin is a critical intermediary in this process, because reducing striatin levels with small interfering RNA (siRNA) technology prevented the rise in pERK levels. In contrast, reducing striatin did not significantly affect 2 well-characterized genomic responses to ALDO. Down-regulation of striatin with siRNA produced similar effects on estrogen's actions, reducing nongenomic, but not some genomic, actions. ALDO, but not estrogen, increased striatin levels. When endothelial cells were pretreated with ALDO, the rapid/nongenomic response to estrogen on phosphorylated endothelial nitric oxide synthase (peNOS) was enhanced and accelerated significantly. Importantly, pretreatment with estrogen did not enhance ALDO's nongenomic response on pERK. In conclusion, our results indicate that striatin is a novel mediator for both ALDO's and estrogen's rapid and nongenomic mechanisms of action on pERK and phosphorylated eNOS, respectively, thereby suggesting a unique level of interactions between the mineralocorticoid receptor and the estrogen receptor in the cardiovascular system.

Figures

Figure 1.

ALDO increases pERK levels. A, EA.hy926 cells were stimulated with different concentrations of ALDO for 15 minutes. Protein levels were measured by Western blot analysis; *, P < .05 compared with time point 0. The minimum ALDO concentration (10−8 mol/L) produced a 40% increase in the pERK/ERK ratio. B, ALDO (10−8 mol/L) leads to a rapid increase in pERK/ERK ratio with maximum stimulation at 15 minutes in EA.hy926 cells. Preincubation with 10−6 mol/L canrenoic acid (MRA), an MRA, prevents the increase in the pERK/ERK ratio, suggesting that ALDO is acting through the MR. Values shown in the graph were calculated comparing the ALDO (A)-treated cells with the respective vehicle (V) and plotted as mean ± SEM.

Figure 2.

Striatin is essential in mediating MR-dependent rapid responses to ALDO. A, Untransfected EA.hy926 cells show an increase in the pERK/ERK ratio, after incubation with ALDO (10−8 mol/L) for 15 minutes (*, P < .05 vs control; **, P < .05 vs ALDO). B, The ALDO-mediated rapid increase in pERK levels is abolished when the EA.hy926 cells are transfected with striatin siRNA (*, P < .01 vs control). C, Transfection with scrambled siRNA does not block the ALDO-induced increase in the pERK/ERK ratio (*, P < .05 vs control; **, P < .05 vs ALDO). D, ROS production was measured as described in Materials and Methods, in the presence (black bars) or absence (vehicle [white bars]) of ALDO (10−8 mol/L) for 60 minutes at 37°C. Untransfected and scrambled siRNA-transfected EA.hy926 cells show an increase in ROS production after incubation with ALDO (P < .05 vs vehicle-treated cells). This ALDO-mediated increase in ROS production is abolished when EA.hy926 cells are transfected with striatin siRNA (*, P < vs vehicle-treated cells; **, P < .01 vs untransfected ALDO-treated cells). Results shown are the averages ± SE of n = 9 independent experiments. E, Transfection of EA.hy926 cells with striatin siRNA but not scrambled leads to reduced striatin protein levels. siRNA transfections were performed as described above and proteins resolved by Western blotting. The Western blotting shown is a representative trace of 6 replicates. The membrane blot used to probe for striatin was stripped, washed, and reprobed for β-actin as loading control. RFU, relative fluorescence units.

Figure 3.

Striatin is nonessential in mediating the MR-dependent genomic responses to ALDO. EA.hy926 cells, transfected with striatin siRNA or controls (untransfected), were treated with ALDO (10–8 mol/L for 15 min and 5 h), and mRNA levels of WNK4 (A) and SGK1 (B) were determined. ALDO increased WNK4 (*, P < .05 vs control) and SGK1 mRNA levels (*, P < .001 vs control) and these genomic responses were not blocked by siRNA to striatin.

Figure 4.

Pretreatment with ALDO increases estrogen's nongenomic response. A, EA.hy926 cells were treated with either 10−8 mol/L ALDO or 5 × 10−8 mol/L estrogen for 5 hours. ALDO but not estrogen treatment increases striatin protein levels. B, EA.hy926 cells were pretreated with or without ALDO (10−8 mol/L) for 5 hours. Then, peNOS/eNOS levels in response to estrogen (5 × 10−8 mol/L) were measured at 5, 15, and 30 minutes. Pretreatment with ALDO significantly increased the peak peNOS/eNOS response to estrogen and shifted the response to an earlier time point. C, EA.hy926 cells were transfected with scrambled (left) or striatin siRNA (right) and pretreated with (white bars) or without ALDO (10−8 mol/L) for 5 hours (black and gray bars). Then, peNOS/eNOS levels in response to estrogen (5 × 10−8 mol/L) were measured at 5 minutes (white bars) and 15 minutes (gray bars), relative to vehicle (black bars). (*, P < .05 vs vehicle) D, EA.hy926 cells were pretreated with or without estrogen (5 × 10−8 mol/L) for 5 hours. Then, pERK/ERK levels in response to ALDO (10−8 mol/L) were measured at 5, 15, and 30 minutes. Pretreatment with estrogen did not alter either the magnitude or time course of ALDO's nongenomic response (*, P < .01 vs vehicle).

Figure 5.

Striatin siRNA does not modify genomic responses to estrogen. EA.hy926 cells, transfected with striatin siRNA or controls (untransfected), were treated with estrogen (5 × 10−8 mol/L), and mRNA levels of PTGIS (A) and PTGS1 (B) were determined. Results show no significant difference in the genomic response after 5 hours of estrogen stimulation in the transfected and control cells (*, P < .05 vs vehicle).

Figure 6.

Effect of ALDO and estrogen incubation in early passages of mouse aortic ECs. A, ALDO (10−8 mol/L) leads to a rapid increase in pERK/ERK ratio. B, Estrogen (5 × 10−8 mol/L) leads to a rapid increase in peNOS/eNOS ratio. C, ECs were pretreated with or without ALDO (10−8 mol/L) for 5 hours. Then, peNOS/eNOS levels in response to estrogen (5 × 10−8 mol/L) were measured at various times. Pretreatment with ALDO resulted in a more rapid estrogen-mediated increase in peNOS/eNOS ratio.

Figure 7.

MR, CAV1, and striatin coprecipitate. A, Western blot showing the presence of CAV1 in ECs, EA.hy926 cells (EAhy) and mouse heart. B–G, IP studies. Striatin coprecipitates with CAV1 (B); CAV1 coprecipitates with striatin, input represents the starting material before IP (C); MR coprecipitates with CAV1 (D); and CAV1 coprecipitates with MR (E). F and G, In heart tissue from CAV1 wild-type (WT) and KO animals, MR only interacts with striatin in WT animals. H, MR coprecipitation with striatin, water, or irrelevant IgG, vs unprecipitated striatin in EA.hy926 cells and mouse heart. WB, Western blot.