PDE5 inhibitor efficacy is estrogen dependent in female heart disease

Abstract

Inhibition of cGMP-specific phosphodiesterase 5 (PDE5) ameliorates pathological cardiac remodeling and has been gaining attention as a potential therapy for heart failure. Despite promising results in males, the efficacy of the PDE5 inhibitor sildenafil in female cardiac pathologies has not been determined and might be affected by estrogen levels, given the hormone's involvement in cGMP synthesis. Here, we determined that the heart-protective effect of sildenafil in female mice depends on the presence of estrogen via a mechanism that involves myocyte eNOS-dependent cGMP synthesis and the cGMP-dependent protein kinase Iα (PKGIα). Sildenafil treatment failed to exert antiremodeling properties in female pathological hearts from Gαq-overexpressing or pressure-overloaded mice after ovary removal; however, estrogen replacement restored the effectiveness of sildenafil in these animals. In females, sildenafil-elicited myocardial PKG activity required estrogen, which stimulated tonic cardiomyocyte cGMP synthesis via an eNOS/soluble guanylate cyclase pathway. In contrast, eNOS activation, cGMP synthesis, and sildenafil efficacy were not estrogen dependent in male hearts. Estrogen and sildenafil had no impact on pressure-overloaded hearts from animals expressing dysfunctional PKGIα, indicating that PKGIα mediates antiremodeling effects. These results support the importance of sex differences in the use of PDE5 inhibitors for treating heart disease and the critical role of estrogen status when these agents are used in females.

Figure 1. Estrogen dependence of sildenafil efficacy in female failing hearts (Gq overexpressors).

(A) FS echocardiographic changes before and after 2 weeks of vehicle or sildenafil treatment in wild-type and Gq-overexpressing (Gq/oe) female mice subjected to sham operation [non-OVX, OVX(–)], OVX [OVX(+)], or OVX with estrogen replacement [OVX(+) + E2] (n = 7–8 per group). *P < 0.05 versus the 2-week vehicle treatment group. (B) Summary data for FS increased after 2 weeks of sildenafil (SIL) treatment (n = 7–8 per group). *P < 0.05 versus non-OVX group; ^†P < 0.05 versus OVX group. (C) Myocardial PKG activity. (D) PKCα activity. (E) Calcineurin (Cn) protein expression was normalized to GAPDH (n = 4–7 per group). *P < 0.05 versus the wild-type groups in non-OVX or OVX mice; ^†P < 0.05 versus the vehicle-without-E2 groups in non-OVX or OVX Gq/oe mice; ^‡P < 0.05 versus the other groups among OVX Gq/oe mice; ^§P < 0.05 versus the other groups among OVX mice. P values shown are for interactions between E2 and SIL treatments (E2 × SIL); 2-way ANOVA.

Figure 2. Estrogen dependence of sildenafil efficacy in female pressure-overloaded hearts and the essential role of PKGIα.

Sildenafil response in OVX wild-type hearts exposed to 2 weeks of TAC with or without E2 replacement (A–D). (A) FS. (B) Myocardial BNP (Nppb) mRNA expression. (C) Heart weight (HW) normalized to tibia length (TL). (D) Myocardial PKG activity (n = 4–8 per group). Sildenafil response in OVX PKGIα-LZM hearts exposed to 2 weeks of TAC with or without E2 replacement (E–G). (E) FS. (F) Nppb mRNA expression. (G) HW/TL (n = 4–9 per group). *P < 0.05 versus the sham group; ^†P < 0.05 versus the vehicle-without-E2 group; ^‡P < 0.05 versus the other groups in TAC; ^§P < 0.05 versus all other groups; P values shown are for interactions between E2 and SIL treatments; 2-way ANOVA.

Figure 3. cGMP production by estrogen coupled to the eNOS/sGC pathway as an upstream regulatory mechanism.

(A) PKGIα protein expression normalized to GAPDH. (B) PDE5 activity in Gq/oe female hearts subjected to sham operation (non-OVX), OVX, and OVX with E2 replacement (n = 5–8 per group). (C) Cardiac myocyte cGMP synthesis in OVX Gq/oe hearts. Isolated cardiac myocytes were treated with E2 (1 nM), E2 plus 10 μM ODQ sGC inhibitor, E2 plus 100 μg/ml HS142-1 (ANP receptor inhibitor), ODQ alone, or HS142-1 alone (n = 3–4 per group). (D) Myocardial PKG activity in OVX hearts lacking eNOS (n = 5–6 per group) or lacking NPRA (n = 4–5 per group) exposed to 2 weeks of TAC, with or without E2 replacement, as well as wild-type controls. *P < 0.05; P values shown are for interactions between E2 and TAC; 2-way ANOVA.

Figure 4. Sex differences in sildenafil responses and the mechanism involving eNOS-cGMP synthesis and PKG activation in Gq/oe hearts.

(A) cGMP synthesis in cardiac myocytes isolated from male Gq/oe hearts, which were treated with 1 nM E2, E2 plus 10 μM ODQ, E2 plus 100 μg/ml HS142-1, ODQ alone, or HS142-1 alone (n = 3–4 per group). *P < 0.05. Sildenafil response as indicated by (B) FS and (C) myocardial PKG activity in male Gq/oe hearts with or without E2 treatment (n = 5–10 per group). *P < 0.05 versus the wild-type group; ^†P < 0.05 versus the vehicle-without-E2 group in Gq/oe mice; ^‡P < 0.05 versus all other groups. (D) Myocyte PKG activity in response to acute sildenafil treatment for 10 minutes in cardiac myocytes isolated from Gq/oe males and OVX females with or without E2 treatment. Data were normalized to wild-type males. n = 5–9 per group. ^†P < 0.05 versus the vehicle-without-E2 group in Gq/oe mice; ^‡P < 0.05 versus all other groups. P values shown are for interactions between E2 and SIL treatments; 2-way ANOVA. (E) Sex differences in phosphorylated eNOS at serine 1177/total expression for eNOS (p/t eNOS) and (F) NOS activity in response to cardiac Gq overexpression (n = 6–9 per group). *P < 0.05; P values shown are for interactions between sexes and genotypes; 2-way ANOVA.