|Home | About | Journals | Submit | Contact Us | Français|
To test the efficacy of phosphodiesterase type-5 (PDE5A) inhibition for treating advanced hypertrophy/remodeling due to pressure-overload, and to elucidate cellular and molecular mechanisms for this response.
Sildenafil (SIL) inhibits cyclic GMP-specific PDE5A and can blunt the evolution of cardiac hypertrophy and dysfunction in mice subjected to pressure-overload. Whether and how it ameliorates more established advanced disease and dysfunction is unknown.
Mice were subjected to transverse aortic constriction (TAC) for 3 weeks to establish hypertrophy/dilation, and subsequently treated with SIL (100 mg/kg/day) or placebo for 6-weeks of additional TAC.
SIL arrested further progressive chamber dilation, dysfunction, fibrosis, and molecular remodeling, increasing myocardial protein kinase G activity. Isolated myocytes from TAC-SIL hearts displayed greater sarcomere shortening and relaxation, and enhanced Ca2+ transients and decay compared to non-treated TAC hearts. SIL treatment restored gene and protein expression of sarcoplasmic reticulum Ca2+ uptake ATPase (SERCA2a) phospholamban (PLB), and increased PLB phosphorylation (S16) – consistent with improved calcium handling. Both the phosphatase calcineurin (Cn) and protein kinase C-α (PKCα) can lower pPLB and depress myocyte calcium cycling. Cn expression and PKCa activation (outer membrane translocation) were enhanced by chronic TAC, and reduced by SIL treatment. PKCδ and PKCε expression also rose with TAC but were unaltered by SIL treatment.
SIL treatment applied to well established hypertrophic cardiac disease can prevent further cardiac and myocyte dysfunction and progressive remodeling. This is associated with improved calcium cycling, and reduction of calcineurin and PKCα activation may be important to this improvement.
Heart failure is a leading cause of morbidity and mortality worldwide, commanding a large share of health care resources, and despite recent advances, better treatment options remain sorely needed(1). Hypertension and corresponding hypertrophic heart disease are major risk factors for heart failure, and the therapeutic suppression of hypertrophy is associated with improved mortality(2;3). To date, treatments designed to blunt hypertrophic remodeling and related dysfunction have largely focused on reducing vascular load and neuro-hormone stimulation. However, intracellular pathways are also now being targeted which may provide potent benefits by interfering with strategic nodes in more distal signaling processes(1;4).
An intriguing approach that first arose somewhat as a surprise is the inhibition of cGMP-specific phosphodiesterase 5A (PDE5A) by compounds such as sildenafil. Such drugs are widely used to treat erectile dysfunction, and long thought to have little role in the heart. However, growing evidence supports cardiac expression and functionality of PDE5A(5) physiologic effects from its inhibitors. For example, PDE5A inhibition can improve ischemic cardiomyopathy (6), doxorubicin toxicity(7) and pressure-overload induced hypertrophy(8), all thought coupled to enhanced protein kinase G activation(9;10). Clinical studies of sildenafil in heart failure patients have reported improved exercise capacity coupled to reduced pulmonary vascular resistance and better endothelial function(11;12). Based on such findings, the NIH is to initiate a multi-center clinical trial of sildenafil for the treatment of heart failure with a preserved ejection fraction.
We previously reported that sildenafil exerts anti-hypertrophic effects against pressure-overload when the treatment is initiated at the onset of the stress or shortly (1-week) thereafter(8). In the latter case, hearts first developed concentric hypertrophy with no dilation, and then improved with treatment. However, it is unknown whether sildenafil can ameliorate more advanced disease, a non-trivial question as PDE5A modulation appears to target nitric oxide-derived cGMP(9;10;13) and nitric oxide synthase (NOS) activity declines with sustained pressure-overload(14). Furthermore, the mechanisms of sildenafil-improved cardiac function in pressure-overload hearts(8) and whether they apply to more advanced disease remain unknown. Accordingly, the present study tested the efficacy of delayed sildenafil treatment of hearts subjected to pressure-overload. We show evidence of chamber, cellular, and molecular benefits of PDE5A inhibition in advanced disease, and reveal novel mechanisms for functional improvement related to calcium cycling and its molecular modulation.
Male C57Bl/6 mice (age 9–12 wks; Jackson Labs, Bar Harbor, ME) were used. Pressure-overload was produced by transverse aortic constriction (TAC) (8), with shams undergoing the same operation without aortic constriction. Oral sildenafil (100 mg/kg/d) was provided in soft diet (Bioserve, MA)(8) starting on week 4 following TAC and continuing for 5 additional weeks. While this dose is high for humans, the mouse metabolizes sildenafil at a higher rate(15), and this dose yields a free plasma concentration of 10–15 nM, within the specific and therapeutic range for PDE5A(8). Control data for chronic sildenafil only have been previously reported(8), and effects found to be negligible. The Johns Hopkins University Animal Care and Use Committee approved the protocol.
Cardiac function was assessed by transthoracic echocardiography in conscious mice (Sequoia C256, Siemens, NY) using a 15 MHz linear-array transducer, and by invasive pressure-volume analysis using the conductance catheter method (SPR-839, Millar Instruments, Inc., Houston, TX). The catheter was inserted via the LV apex in open-chest, anesthetized mice, and positioned along the long-axis as described(8)
Formalin (10%) fixed paraffin embedded LV myocardium was sectioned (5 μm) and stained with hematoxylin/eosin or picrosirius red. Myocyte diameter and interstitial and peri-vascular collagen fraction were determined by digital image analysis (Adobe Photoshop 7.0, NIH Image J), with the observer blinded as to tissue source. Four different hearts in each group, with five separate fields of cells (total 50–70 cells for each heart) were analyzed.
Sarcomere shortening (Myocam, IonOptix, Milton, MA) and whole-cell Ca2+ transient (Fura-2 AM μmol/L; Invitrogen, Carlsbad, CA) was assessed in freshly isolated adult cardiomyocytes (4 hearts/group) at 27°C using 0.5 Hz field stimulation (15–20 cells from each heart) as previously described(10). Though lower temperatures can depress adrenergic modulation, we have shown marked concordance of adrenergic regulation by PDE5 inhibitors under various conditions between such in vitro conditions and in vivo hearts(9;10). Data at rest and after isoproterenol (ISO; 10 nmol/L) stimulation were obtained.
Protein and mRNA isolates were prepared from LV tissue flash frozen in liquid nitrogen, and expression assessed using standard techniques (see online-supplemental methods).
Data are presented as mean ± SEM. Differences between groups were assessed by 1- or 2-way analysis of variance (with or without repeated measures, as required) followed by a Tukey’s multiple comparisons test. In instances where within-group variance differed substantially, a non- parametric Kruskal-Wallis test and Bonferoni correction was used.
After 3-wk TAC, hearts had a +135% rise in LV mass, chamber end-systolic (+91%) and end-diastolic (+10%) dimensions, and reduced fractional shortening (−42%; Fig 1a). Subsequent treatment with sildenafil fully arrested progressive remodeling whereas control hearts further dilated and hypertrophied after 9 weeks TAC. Post-mortem analysis confirmed both heart and lung weight, normalized to tibia length, was lower with sildenafil treatment (Fig 1b). Myocyte cross sectional dimension and interstitial and peri-vascular fibrosis was also reduced in SIL-treated myocardium (Fig 1c).
A- and B-type natriuretic peptide gene expression rose markedly after 3-wks-TAC. Both increased 30–50% after 9-wks-TAC in non-treated hearts but remained at 3-wk levels in those receiving sildenafil (Fig 2a). β-myosin heavy chain also increased, but this was not reduced in sildenafil treated hearts.
We next assessed the activation of cGMP-stimulated protein kinase G (PKG) in hearts with or without delayed sildenafil treatment. Enzyme activity assessed by S-239-VASP phosphorylation and in vitro kinase assay both showed increases after 9wk-TAC that were further enhanced in SIL treated animals (Fig 2b). TAC led to increased PKG-1α (primary cardiac isoform) protein expression (Fig 2c), but this declined to normal levels with SIL treatment, supporting post-translational (cGMP-stimulation) mechanisms in this setting. PDE5A protein expression was unaltered among the various conditions.
Figure 3A shows the results of LV pressure-volume analyses. After 3-wks-TAC, hearts dilated and had increased end-systolic elastance (Ees) and contractility (dP/dtmax/IP and maximal power index), as reported in humans with hypertrophic heart disease(16). More sustained TAC reduced contractility and prolonged relaxation in untreated hearts but this was prevented by SIL.
To identify mechanisms for improved cardiac function, LV myocytes were studied (Fig 3b). Compared to sham controls, 9-wk-TAC myocytes had depressed sarcomere shortening and prolonged relaxation, and SIL treatment restored both to control levels. These changes were accompanied by improved calcium handling. The peak Ca2+-transient was not significantly altered after 9-wk-TAC but the time for 50% decline prolonged. The latter improved towards control values with SIL treatment, while peak amplitude rose. Contraction, relaxation and Ca2+-transients all improved similarly with ISO in each group (p-value for interaction by 2-Way ANOVA ranged 0.2–0.94); thus, this cascade was not differentially affected by SIL treatment.
Since CA2+-cycling improved in SIL treated TAC hearts, we examined sarcoplasmic reticulum (SR) handling proteins that might be modified by the therapy. Gene expression of SR-Ca2+ ATPase (SERCA2a) and phospholamban declined after 3-wks-TAC (Fig 4a) and this persisted after 9-wks-TAC. Protein expression was declined at this time as well. SIL treatment increased both SERCA2a and PLB expression, and enhanced PLB s-23 phosphorylation (increasing Ca2+-uptake), in comparison to non-treated TAC.
Increased PLB phosphorylation by SIL is consistent with enhanced PKG activity, but it could also result from suppressed phosphatase activity. One mechanism involves calcineurin (Cn) which inhibits I-1 activity resulting in the dis-inhibition of the phosphatase PP1 to lower PLB phosphorylation(17;18). Cn protein expression and activity (indexed by RCAN-1 mRNA) rose after 3-wks-TAC by 3.2 and 2.5-fold, respectively (both p<0.01, on-line supplemental Figure 1), and remained elevated at 9-wks-TAC in non-treated hearts but declined with SIL (Fig. 4c). Another mechanism for I-1 inactivation is its phosphorlyation by PKCα at S67 and T75 which also enhances PP1 activity and thus PLB de-phosphorylation (19). PKCα (and δ,ε isoforms) protein expression rose markedly after 9-wks-TAC (Fig 4d), and SIL treatment did not reduce them. However, PKC-α activity is coupled to its translocation to the outer membrane, and this was observed after both 3 and 9-wks of TAC (Fig 5a). PKC-α returned to a diffuse distribution consistent with reduced activation following SIL treatment. Confocal analysis was confirmed by cell fractionation/immunoblot analysis (Fig 5b). To test if SIL directly targeted PKC-α activation, freshly isolated cardiomyocytes were incubated with the PKC activator PMA. PMA induced rapid PKC-α membrane translocation (Fig. 5c) but which SIL co-incubation did not suppress, suggesting the in vivo effect was more likely indirect.
Cardiac hypertrophy and attendant myocardial remodeling and myocyte and chamber dysfunction remain major causes of morbidity and mortality worldwide, and new approaches to combat this pathophysiology are needed. In a prior study, we first showed that PDE5A inhibition coupled to activation of PKG may offer a novel approach to treating this disorder(8). The present results substantially extend this finding. First therapy was initiated only after the hypertrophic disease process was far more established, yet improvements in function, remodeling, and molecular signaling were achieved. Second, isolated myocytes were studied revealing enhanced myocyte contraction/relaxation and Ca2+ handling under both rest and β-AR stimulated conditions. Third, we extended prior mechanistic analysis, showing improvement of SR calcium handling proteins coupled with suppression of both Cn and PKC-α activation. These findings further support a translational potential for PDE5A inhibitors in established hypertrophic heart disease.
Though the potential for cGMP/PKG signaling to suppress cardiac hypertrophy has been recognized for some time, it has been difficult to translate into an effective therapy. Prior studies have focused on increasing cGMP synthesis via natriuretic peptides or nitric oxide, but this remains compromised by peripheral vasodilation and tachyphylaxis in part due to feedback inhibition by phosphodiesterases(20;21). Even in genetically engineered animals with NP or NOS pathways modulated(22;23), TAC-induced hypertrophy changes have been modest, and no study has examined a situation where the disease was already well established.
Suppression of cGMP hydrolysis provides an alternative approach. Of three PDE species identified in heart to date(5), two are dual substrate (PDE1 and PDE2), the former requiring Ca2+-calmodulin activation and the latter also acting as a cGMP stimulated cAMP hydrolytic enzyme. Their role in physiologic cardiac cGMP regulation remains largely unknown. PDE5a was the first selective cGMP-PDE discovered and remains the best characterized(5). Though first thought to have little role in the heart, growing evidence supports its regulation of a localized cGMP pool that can potently modulate cardiac stress responses(5–8), and the current data strongly supports this further.
cGMP hydrolysis by PDE5A is compartmentalized and appears to particularly target NOS-generated cGMP(9;10;24). However, sustained pressure-overload results in functional NOS uncoupling, reducing its synthesis of NO and increasing its generation of superoxide(14). The fact that SIL remains effective in such a setting is important and may relate to observations that PDE5a inhibition reduces NO-ROS interaction decline in nitrotyrosine and improves NOS coupling and activity in TAC hearts(13). The mechanism for this is presently under study. It is also worth noting that therapies targeting NO-sGMP-PKG do not necessarily lead to the same end-modulation (i.e. PKG activation). For example, hearts exposed to sustained pressure-overload but then treated with tetrahydrobiopterin, an essential cofactor required for NOS coupling and thus normal function, also display suppression of progressive remodeling, dysfunction, hypertrophy and fibrosis, yet unlike SIL, do not have heightened PKG activity(25).
Acute PDE5A inhibition minimally alters resting heart or myocyte function as basal cGMP-cyclase activity is low(9;10); however, it can substantially blunt β-adrenergic stimulation which co-activates the cyclase(10;26). Likewise, chronic SIL (same dose used in the present study) also minimally affects the normal heart(8), yet improves function in hearts and myocytes subjected to sustained pressure-overload. The apparent paradox between acute negative yet chronic positive contractility responses shares some similarity to β-blockade. Acute β-blockade is negatively inotropic but this effect depends upon the level of adrenergic tone, whereas chronic blockade in a stressed/failing heart enhances contraction likely by suppressing sustained catecholamine cytotoxicity(27). However, β-blockers are far less potent in suppressing pressure-overload hypertrophy and their utility in hypertensive heart disease has been questioned(28). By contrast, PDE5A inhibitors target more distal signaling and suppress hypertrophy and maladaptive molecular changes triggered by mechanical and neurohumoral stress even without altering blood pressure. By blocking pathways that otherwise depress Ca2+ handling, myocyte function, and remodeling, the net chronic effect positively impact contractility.
Our results contrast to those of a recent study performed in chronic right ventricular hypertrophied rat hearts, where PDE5A inhibition acutely augmented contractility linked to a rise in cAMP (not cGMP), and activation of PKA (not PKG) (29). In this study, myocyte shortening was equally enhanced by isoproterenol, sildenafil, or their combination, suggesting the same pathway was activated by both. RV hypertrophy reduced PKG activity (and pVASP) and increased PDE5A expression. In contrast, we found ISO augmented LV myocyte shortening similarly in mice subjected to TAC±SIL, hypertrophy enhanced PKG activity (and pVASP) and sildenafil raised it further, and PDE5a expression was unchanged. In our prior study, myocardial cAMP was also similar in hearts subjected to chronic TAC±SIL(8). The difference may lie between chronic versus acute interventions, right versus left ventricle, or conceivably species. With regard to the latter, PDE5A inhibitors blunt β-adrenergic stimulation in humans(26) similar to their effect in mice(10).
Sildenafil acutely depresses ISO-stimulated contraction without altering the whole cell Ca2+ transient(10). This may reflect a balance between a PKG-mediated decline in L-type Ca2+ current(30), reduced myofilament sensitivity(31), and PLB phosphorylation(32). With chronic pressure-overload, however, other regulators appear involved. Reduced PLB phosphorylation is found in experimental and human heart failure(32), and both Cn and PKC-α are thought play an role. PKC-α is the dominant isozyme in heart and translocates to the outer sarcolemmal membrane upon stimulation(33). It phosphorylates and thus inhibits I-1 (inhibitor of the phosphatase PP1), reducing PLB phosphorylation. Genetic(19) and pharmacologic(34) blockade of PKC-α improves function, though its impact on hypertrophy is less marked. While SIL did not alter PKC-α (or other PKC isoform) protein expression, it shift it away from the outer membrane consistent with de-activation. I-1 can also be inhibited by Cn via de-phosphorlyating sites normally stimulated by PKA (Thr-35) (17;35). Future studies are needed to directly prove the role of I-1 modulation to SIL-modified cardiac function.
Delayed sildenafil treatment inhibits the progression of hypertrophy, fibrosis, and chamber remodeling, and improves basal and β-stimulated contractility and relaxation in a model of established pressure-overload induced hypertrophic heart disease. Functional improvement is coupled to enhanced Ca2+ handling, and increased expression of SR Ca2+ uptake proteins and PLB phosphorylation likely contribute. The capacity of SIL to block maladaptive remodeling even when administered later in the disease process provides important experimental support for currently planned clinical studies, where disease will already be present.
Sources of Funding: This study was supported by National Institute of Health Grants: HL-59408; HL-07227; HL- 084946, Abraham and Virginia Weiss Professorship, Peter Belfer Laboratory, and and Robert Kubicki Research Fund (DAK), fellowship grant from Daiichi-Sankyo Inc. (TN), T32-HL-07227 (MZ), American Heart Association SDG Award and HL-084946 (ET), and a fellowship grant from Japan Heart Foundation (NK).
Disclosures: The authors have no disclosures to report.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.