This report describes studies designed to study roles of PDE3A and PDE3B in regulation of myocardial function and energy homeostasis. Myocardial function: In human and mouse heart, cAMP stimulates myocardial contractility by increasing protein kinase A (PKA)-induced phosphorylation of membrane-bound substrates involved in intracellular Ca2+ cycling and excitation/contraction coupling. Using PDE3A and PDE3B KO mice, we found that PDE3A, not PDE3B, regulates basal contractility, and that inhibition of PDE3A, not PDE3B, mediates the inotropic effects of the PDE3 inhibitor, milrinone (Primacor) (Circ Res 112:289-97, 2013). PDE3A regulates basal contractility and cAMP-mediated Ca++ uptake into the SR as a component of a SERCA2 regulatory complex or signalosome which contains AKAP18, PKA, SERCA2, phospholamban, and PP2A. Ongoing work (Faiyaz Ahmad, Staff Scientist) indicates that in human heart, PDE3A is localized with SERCA2, PLB and AKAP18 on sarcomere Z-line bands, and that a similar AKAP18/SERCA2/ PDE3A-containing signalosome regulates SERCA2 and Ca++ uptake in human SR. In these preparations, signalsome assembly/formation is enhanced by phosphorylation, and experiments with recombinant proteins indicate that rPDE3A may directly interact with rSERCA2 and rAKAP18. Three PDE3A isoforms, PDE3A1, PDE3A2 and PDE3A3 are expressed in human heart;they possess identical amino-acid sequences except for deletion of different lengths of the N-terminal region. Collaborative studies (PNAS, revision submitted) demonstrated that in HEK293 cells, rPDE3A1 and rPDE3A2 were differentially phosphorylated at distinct 14-3-3 binding sites by isoproterenol and phorbol ester, respectively. Phosphorylation of rPDE3A1 by PKA and of rPDE3A2 by PKC induced shifts in their elution on gel-filtration chromatography, consistent with their phosphorylation-dedendent incorporation into different regulatory signalosomes. Selective phosphorylation of PDE3A1 and PDE3A2 at alternative sites via different signaling pathways, together with different functional consequences of phosphorylation for each, suggest they are likely to have distinct roles in cyclic nucleotide-mediated signaling in human myocardium. With respect to PDE3B, we demonstrated (Circ Res, in revision) that targeted disruption of PDE3B, but not PDE3A, protected mouse hearts from ischemia reperfusion (I/R)-induced injury in vivo and in vitro, with significantly reduced infarct size. Administration of milrinone, a PDE3 inhibitor, to mice, prior to induction of ischemia, reduced infarct size in WT and PDE3A KO mice, but did not further increase protection in PDE3B KO mice. Deletion of PDE3B protected Langendorff-perfused hearts from I/R injury, most likely via enhanced opening of mitoKCa channels, less ROS production, and reduced Ca2+-induced opening of the mitochondrial permeability transition pore (mPTP) in PDE3B KO mitochondria. The mechanism(s) for cardioprotection may involve activation of PI3K/Akt/GSK-3 signaling pathways and cAMP/PKA-induced assembly of ICEF (Ischemia-induced caveolin-enriched fractions), in which various cardioprotective molecules accumulate, resulting in functional preconditioning in PDE3B KO hearts. ICEF are buoyant, caveolae-like fractions, separated from crude mitochondrial fractions by discontinuous sucrose gradient centrifugation. ICEF contain protein components of membrane repair complexes (dysferlin, annexin A2, caveolin-3 and TRIM72), calcium signaling proteins, and other proteins associated with cardioprotection from I/R injury. Protective effects associated with inhibition/deletion of PDE3B may reflect roles of PDE3A and PDE3B in regulating, at distinct subcellular sites, compartmentalization of specific cAMP-signaling pathways, since cryo-immunogold electron microscopy of ventricular muscle revealed that PDE3A was localized with SERCA on SR membranes, whereas PDE3B was localized with caveolin-3 on T-tubule membranes along the Z-line and within the sarcomere I-band, in regions where mitochondria are in close contact with both SR and T-tubules. Furthermore, in PDE3B KO hearts, ICEF, with their cardioprotective molecules, may be delivered to mitochondria via T-tubules, since analysis of heart electron micrographs demonstrated that contacts between T-tubules and mitochondria were increased in PDE3B KO hearts, compared to WT. PDE3 inhibitors (e.g. cilostazol (Pletal)) are in common use for treating intermittent claudication, a peripheral vascular disease, although earlier clinical trials with heart failure subjects demonstrated that chronic inhibition of PDE3 with milrinone (which increased contractility) increased the incidence of ventricular arrhythmias and mortality. Existing PDE3 inhibitors, however, have little selectivity for PDE3A versus PDE3B isoforms, whose catalytic domains are similar, and no selectivity for individual PDE3A isoforms, which possess identical catalytic domains. Isoform-selective targeting may increase contractility in failing hearts without increasing mortality, thus providing a novel route for developing therapeutics. Blocking the integration of PDE3A isoforms into different signalosomes, either by blocking PDE3A phosphorylation or blocking its interactions with constituents of the signalosomes, may be a another way of targeting PDE3A in a specific microdomain to produce inotropic actions without the adverse consequences that accompany diffuse increases in intracellular cAMP. Our findings also suggest that inhibition of cardiac PDE3B, not PDE3A, might account for reported cardioprotective effects of cilostazol from experimental I/R injury. Furthermore, PDE3B-selective inhibitors might provide benefit in heart transplant patients and heart failure patients, by limiting I/R damage. In this regard, a current collaborative project with Dr. Peter Backx (U Toronto, Canada) will study the clinical course and pathophysiological sequellae of transaortic constriction in WT and PDE3A KO and PDE3B KO mice, and effects of milrinone in these groups. PDE3B regulates energy homeostasis: PDE3B regulates energy metabolism (J Clin Invest 116:3240-3251, 2006), and recent studies (Endocrinology 154:3152-67, 2013) indicate that, in PDE3B KO mice (C57Bl6 background), white adipose tissue (WAT) assumes phenotypic characteristics of brown adipose tissue (BAT). The WAT/BAT phenotypic conversion was markedly enhaced by the Beta3 receptor agonist CL316243, and mediated, perhaps, by cAMP-induced differentiation of prostaglandin-responsive progenitor cells in KO WAT stromal vascular fractions into functional brown adipocytes. The appearance of BAT-like characteristics was accompanied by an increase in oxygen consumption and induction of genes involved in BAT recruitment (cyclooxygenase-2 (COX-2) and elongation of very long chain fatty acids 3 (Elovl3)), and its thermogenic program (PGC-1, uncoupling protein 1 (Ucp1)). Unpublished studies indicate that in SvJ129 PDE3B KO mice, WAT also assumes phenotypic characteristics of BAT, without administration of CL316243 and without induction of COX-2, suggesting critical influences of genetic background on development of the BAT phenotype. In SvJ129 PDE3B KO WAT expression of pro-inflammatory markers is reduced, compared to WT, as are components of the NLRP3 inflammasome (activation of the NLRP3 inflammasome may be related to insulin resistance and obesity-related inflammation).These studies are important, since reducing inflammation in WAT and inducing WAT to assume characteristics of BAT is viewed as potential treatment for obesity and related disorders. Furthermore, these studies with PDE3B KO mice complement our collaborative study (J. Chung, NHLBI) which demonstrated that beneficial/therapeutic effects of resveratrol on energy metabolism may be mediated by inhibition of PDEs, including PDE3 and PDE4 (Cell 148:421-433, 2012).
Showing the most recent 10 out of 32 publications
