Development of a fission yeast-based high-throughput screen to identify chemical regulators of cAMP phosphodiesterases

Cyclic nucleotide phosphodiesterases (PDEs) comprise a superfamily of enzymes that serve as drug targets in many human diseases. There is a continuing need to identify high-specificity inhibitors that affect individual PDE families or even subtypes within a single family. The authors describe a fission yeast-based high-throughput screen to detect inhibitors of heterologously expressed adenosine 3',5'-cyclic monophosphate (cAMP) PDEs. The utility of this system is demonstrated by the construction and characterization of strains that express mammalian PDE2A, PDE4A, PDE4B, and PDE8A and respond appropriately to known PDE2A and PDE4 inhibitors. High-throughput screens of 2 bioactive compound libraries for PDE inhibitors using strains expressing PDE2A, PDE4A, PDE4B, and the yeast PDE Cgs2 identified known PDE inhibitors and members of compound classes associated with PDE inhibition. The authors verified that the furanocoumarin imperatorin is a PDE4 inhibitor based on its ability to produce a PDE4-specific elevation of cAMP levels. This platform can be used to identify PDE activators, as well as genes encoding PDE regulators, which could serve as targets for future drug screens.     

Role of phosphodiesterase 2 in growth and invasion of human malignant melanoma cells

Cyclic nucleotide phosphodiesterases (PDEs) regulate the intracellular concentrations and effects of adenosine 3′,5′-cyclic monophosphate (cAMP) and guanosine 3′,5′-cyclic monophosphate (cGMP). The role of PDEs in malignant tumor cells is still uncertain. The role of PDEs, especially PDE2, in human malignant melanoma PMP cell line was examined in this study. In PMP cells, 8-bromo-cAMP, a cAMP analog, inhibited cell growth and invasion. However, 8-bromo-cGMP, a cGMP analog, had little or no effect. PDE2 and PDE4, but not PDE3, were expressed in PMP cells. Growth and invasion of PMP cells were inhibited by erythro-9-(2-hydroxy-3-nonyl) adenine (EHNA), a specific PDE2 inhibitor, but not by rolipram, a specific PDE4 inhibitor. Moreover, cell growth and invasion were inhibited by transfection of small interfering RNAs (siRNAs) specific for PDE2A and a catalytically-dead mutant of PDE2A. After treating cells with EHNA or rolipram, intracellular cAMP concentrations were increased. Growth and invasion were stimulated by PKA14-22, a PKA inhibitor, and inhibited by N⁶-benzoyl-c AMP, a PKA specific cAMP analog, whereas 8-(4-chlorophenylthio)-2′-O-methyl-cAMP, an Epac specific cAMP analog, did not. Invasion, but not growth, was stimulated by A-kinase anchor protein (AKAP) St-Ht31 inhibitory peptide. Based on these results, PDE2 appears to play an important role in growth and invasion of the human malignant melanoma PMP cell line. Selectively suppressing PDE2 might possibly inhibit growth and invasion of other malignant tumor cell lines.     

The Phosphodiesterase 9 Inhibitor PF‐04449613 Promotes Dendritic Spine Formation and Performance Improvement after Motor Learning

The cyclic nucleotide cGMP is an intracellular second messenger with important roles in neuronal functions and animals' behaviors. The phosphodiesterases (PDEs) are a family of enzymes that hydrolyze the second messengers cGMP and cAMP. Inhibition of phosphodiesterase 9 (PDE9), a main isoform of PDEs hydrolyzing cGMP, has been shown to improve learning and memory as well as cognitive function in rodents. However, the role of PDE9 in regulating neuronal structure and function in vivo remains unclear. Here we used in vivo two‐photon microscopy to investigate the effect of a selective PDE9 inhibitor PF‐04449613 on the activity and plasticity of dendritic spines of layer V pyramidal neurons in the mouse primary motor cortex. We found that administration of PF‐04449613 increased calcium activity of dendrites and dendritic spines of layer V pyramidal neurons in mice under resting and running conditions. Chronic treatment of PF‐04449613 over weeks increased dendritic spine formation and elimination under basal conditions. Furthermore, PF‐04449613 treatment over 1–7 days increased the formation and survival of new spines as well as performance improvement after rotarod motor training. Taken together, our studies suggest that elevating the level of cGMP with the PDE9 inhibitor PF‐04449613 increases synaptic calcium activity and learning‐dependent synaptic plasticity, thereby contributing to performance improvement after learning. © 2018 Wiley Periodicals, Inc. Develop Neurobiol 00: 000–000, 2018     

Phosphodiesterase types 3 and 4 regulate the phasic contraction of neonatal rat bladder smooth myocytes via distinct mechanisms

Activation of the cyclic AMP (cAMP) pathway reduces bladder contractility. However, the role of phosphodiesterase (PDE) families in regulating this function is poorly understood. Here, we compared the contractile function of the cAMP hydrolyzing PDEs in neonatal rat bladder smooth myocytes. RT-PCR and Western blotting analysis revealed that several isoforms of PDE1–4 were expressed in neonatal rat bladder. While 8-methoxymethyl-3-isobutyl-1-methylxanthine (a PDE1 inhibitor) and BAY-60-7550 (a PDE2 inhibitor) had no effect on the carbachol-enhanced phasic contractions of bladder strips, cilostamide (Cil, a PDE3 inhibitor) and Ro-20-1724 (Ro, a PDE4 inhibitor) significantly reduced these contractions. This inhibitory effect of Ro was blunted by the PKA inhibitor H-89, while the inhibitory effect of Cil was strongly attenuated by the PKG inhibitor KT 5823. Application of Ro in single bladder smooth myocytes resulted in an increase in Ca^2 + spark frequency but a decrease both in Ca^2 + transients and in sarcoplasmic reticulum (SR) Ca^2 + content. In contrast, Cil had no effect on these events. Furthermore, Ro-induced inhibition of the phasic contractions was significantly blocked by ryanodine and iberiotoxin. Taken together, PDE3 and PDE4 are the main PDE isoforms in maintaining the phasic contractions of bladder smooth myocytes, with PDE4 being functionally more active than PDE3. However, their roles are mediated through different mechanisms.     

Kinetic properties of Ca2+/calmodulin-dependent phosphodiesterase isoforms dictate intracellular cAMP dynamics in response to elevation of cytosolic Ca2+

Multiply regulated adenylyl cyclases (AC) and phosphodiesterases (PDE) can yield complex intracellular cAMP signals. Ca2+-sensitive ACs have received far greater attention than the Ca2+/calmodulin-dependent PDE (PDE1) family in governing intracellular cAMP dynamics in response to changes in the cytosolic Ca2+ concentration ([Ca2+]i). Here, we have stably expressed two isoforms of PDE1, PDE1A2 and PDE1C4, in HEK-293 cells to determine whether they exert different impacts on cellular cAMP. Fractionation and imaging showed that both PDEs occurred mainly in the cytosol. However, PDE1A2 and PDE1C4 differed considerably in their ability to hydrolyze cAMP and in their susceptibility to inhibition by the non-selective PDE inhibitor, IBMX and the PDE1-selective inhibitor, MMX. PDE1A2 had an approximately 30-fold greater Km for cAMP than PDE1C4 and yet was more susceptible to inhibition by IBMX and MMX than was PDE1C4. These differences were mirrored in intact cells when thapsigargin-induced capacitative Ca2+ entry (CCE) activated the PDEs. Mirroring their kinetic properties, PDE1C4 was active at near basal cAMP levels, whereas PDE1A2 required agonist-triggered levels of cAMP, produced in response to stimulation of ACs. The effectiveness of IBMX and MMX to inhibit PDE1A2 and PDE1C4 in functional studies was inversely related to their respective affinities for cAMP. To assess the impact of the two isoforms on cAMP dynamics, real-time cAMP measurements were performed in single cells expressing the two PDE isoforms and a fluorescent Epac-1 cAMP biosensor, in response to CCE. These measurements showed that prostaglandin E1-mediated cAMP production was markedly attenuated in PDE1C4-expressing cells upon induction of CCE and cAMP hydrolysis occurred at a faster rate than in cells expressing PDE1A2 under similar conditions. These results prove that the kinetic properties of PDE isoforms play a major role in determining intracellular cAMP signals in response to physiological elevation of [Ca2+]i and thereby provide a rationale for the utility of diverse PDE1 species.     

Crystal structure of human phosphodiesterase 3B: atomic basis for substrate and inhibitor specificity

Phosphodiesterases (PDEs) are enzymes that modulate cyclic nucleotide signaling and as such are clinical targets for a range of disorders including congestive heart failure, erectile dysfunction, and inflammation. The PDE3 family comprises two highly homologous subtypes expressed in different tissues, and inhibitors of this family have been shown to increase lipolysis in adipocytes. A specific PDE3B (the lipocyte-localized subtype) inhibitor would be a very useful tool to evaluate the effects of PDE3 inhibition on lipolysis and metabolic rate and might become a novel tool for treatment of obesity. We report here the three-dimensional structures of the catalytic domain of human PDE3B in complex with a generic PDE inhibitor and a novel PDE3 selective inhibitor. These structures explain the dual cAMP/cGMP binding capabilities of PDE3, provide the molecular basis for inhibitor specificity, and can supply a valid platform for the design of improved compounds.     

PDE8 Regulates Rapid Teff Cell Adhesion and Proliferation Independent of ICER

Background

Abolishing the inhibitory signal of intracellular cAMP by phosphodiesterases (PDEs) is a prerequisite for effector T (Teff) cell function. While PDE4 plays a prominent role, its control of cAMP levels in Teff cells is not exclusive. T cell activation has been shown to induce PDE8, a PDE isoform with 40- to 100-fold greater affinity for cAMP than PDE4. Thus, we postulated that PDE8 is an important regulator of Teff cell functions.

Methodology/Principal Findings

We found that Teff cells express PDE8 in vivo. Inhibition of PDE8 by the PDE inhibitor dipyridamole (DP) activates cAMP signaling and suppresses two major integrins involved in Teff cell adhesion. Accordingly, DP as well as the novel PDE8-selective inhibitor PF-4957325-00 suppress firm attachment of Teff cells to endothelial cells. Analysis of downstream signaling shows that DP suppresses proliferation and cytokine expression of Teff cells from Crem ^−/− mice lacking the inducible cAMP early repressor (ICER). Importantly, endothelial cells also express PDE8. DP treatment decreases vascular adhesion molecule and chemokine expression, while upregulating the tight junction molecule claudin-5. In vivo, DP reduces CXCL12 gene expression as determined by in situ probing of the mouse microvasculature by cell-selective laser-capture microdissection.

Conclusion/Significance

Collectively, our data identify PDE8 as a novel target for suppression of Teff cell functions, including adhesion to endothelial cells.     

Structural basis for the activity of drugs that inhibit phosphodiesterases

Phosphodiesterases (PDEs) comprise a large family of enzymes that catalyze the hydrolysis of cAMP or cGMP and are implicated in various diseases. We describe the high-resolution crystal structures of the catalytic domains of PDE4B, PDE4D, and PDE5A with ten different inhibitors, including the drug candidates cilomilast and roflumilast, for respiratory diseases. These cocrystal structures reveal a common scheme of inhibitor binding to the PDEs: (i) a hydrophobic clamp formed by highly conserved hydrophobic residues that sandwich the inhibitor in the active site; (ii) hydrogen bonding to an invariant glutamine that controls the orientation of inhibitor binding. A scaffold can be readily identified for any given inhibitor based on the formation of these two types of conserved interactions. These structural insights will enable the design of isoform-selective inhibitors with improved binding affinity and should facilitate the discovery of more potent and selective PDE inhibitors for the treatment of a variety of diseases.     

Phosphodiesterases catalyze hydrolysis of cAMP-bound to regulatory subunit of protein kinase A and mediate signal termination

Although extensive structural and biochemical studies have provided molecular insights into the mechanism of cAMP-dependent activation of protein kinase A (PKA), little is known about signal termination and the role of phosphodiesterases (PDEs) in regulatory feedback. In this study we describe a novel mode of protein kinase A-anchoring protein (AKAP)-independent feedback regulation between a specific PDE, RegA and the PKA regulatory (RIα) subunit, where RIα functions as an activator of PDE catalysis. Our results indicate that RegA, in addition to its well-known role as a PDE for bulk cAMP in solution, is also capable of hydrolyzing cAMP-bound to RIα. Furthermore our results indicate that binding of RIα activates PDE catalysis several fold demonstrating a dual function of RIα, both as an inhibitor of the PKA catalytic (C) subunit and as an activator for PDEs. Deletion mutagenesis has localized the sites of interaction to one of the cAMP-binding domains of RIα and the catalytic PDE domain of RegA whereas amide hydrogen/deuterium exchange mass spectrometry has revealed that the cAMP-binding site (phosphate binding cassette) along with proximal regions important for relaying allosteric changes mediated by cAMP, are important for interactions with the PDE catalytic domain of RegA. These sites of interactions together with measurements of cAMP dissociation rates demonstrate that binding of RegA facilitates dissociation of cAMP followed by hydrolysis of the released cAMP to 5'AMP. cAMP-free RIα generated as an end product remains bound to RegA. The PKA C-subunit then displaces RegA and reassociates with cAMP-free RIα to regenerate the inactive PKA holoenzyme thereby completing the termination step of cAMP signaling. These results reveal a novel mode of regulatory feedback between PDEs and RIα that has important consequences for PKA regulation and cAMP signal termination.     

The calcium/calmodulin-dependent phosphodiesterase PDE1C down-regulates glucose-induced insulin secretion.

To understand the role cAMP phosphodiesterases (PDEs) play in the regulation of insulin secretion, we analyzed cyclic nucleotide PDEs of a pancreatic beta-cell line and used family and isozyme-specific PDE inhibitors to identify the PDEs that counteract glucose-stimulated insulin secretion. We demonstrate the presence of soluble PDE1C, PDE4A and 4D, a cGMP-specific PDE, and of particulate PDE3, activities in betaTC3 insulinoma cells. Selective inhibition of PDE1C, but not of PDE4, augmented glucose-stimulated insulin secretion in a dose-dependent fashion thus demonstrating that PDE1C is the major PDE counteracting glucose-dependent insulin secretion from betaTC3 cells. In pancreatic islets, inhibition of both PDE1C and PDE3 augmented glucose-dependent insulin secretion. The PDE1C of betaTC3 cells is a novel isozyme possessing a K(m) of 0.47 microM for cAMP and 0.25 microM for cGMP. The PDE1C isozyme of betaTC3 cells is sensitive to 8-methoxymethyl isobutylmethylxanthine and zaprinast (IC(50) = 7.5 and 4.5 microM, respectively) and resistant to vinpocetine (IC(50) > 100 microM). Increased responsiveness of PDE1C activity to calcium/calmodulin is evident upon exposure of cells to glucose. Enhanced cAMP degradation by PDE1C, due to increases in its responsiveness to calcium/calmodulin and in intracellular calcium, constitutes a glucose-dependent feedback mechanism for the control of insulin secretion.     

Cyclic AMP control measured in two compartments in HEK293 cells: phosphodiesterase K(M) is more important than phosphodiesterase localization

The intracellular second messenger cyclic AMP (cAMP) is degraded by phosphodiesterases (PDE). The knowledge of individual families and subtypes of PDEs is considerable, but how the different PDEs collaborate in the cell to control a cAMP signal is still not fully understood. In order to investigate compartmentalized cAMP signaling, we have generated a membrane-targeted variant of the cAMP Bioluminiscence Resonance Energy Transfer (BRET) sensor CAMYEL and have compared intracellular cAMP measurements with it to measurements with the cytosolic BRET sensor CAMYEL in HEK293 cells. With these sensors we observed a slightly higher cAMP response to adenylyl cyclase activation at the plasma membrane compared to the cytosol, which is in accordance with earlier results from Fluorescence Resonance Energy Transfer (FRET) sensors. We have analyzed PDE activity in fractionated lysates from HEK293 cells using selective PDE inhibitors and have identified PDE3 and PDE10A as the major membrane-bound PDEs and PDE4 as the major cytosolic PDE. Inhibition of membrane-bound or cytosolic PDEs can potentiate the cAMP response to adenylyl cyclase activation, but we see no significant difference between the potentiation of the cAMP response at the plasma membrane and in cytosol when membrane-bound and cytosolic PDEs are inhibited. When different levels of stimulation were tested, we found that PDEs 3 and 10 are mainly responsible for cAMP degradation at low intracellular cAMP concentrations, whereas PDE4 is more important for control of cAMP at higher concentrations.     

A glutamine switch mechanism for nucleotide selectivity by phosphodiesterases

Phosphodiesterases (PDEs) comprise a family of enzymes that modulate the immune response, inflammation, and memory, among many other functions. There are three types of PDEs: cAMP-specific, cGMP-specific, and dual-specific. Here we describe the mechanism of nucleotide selectivity on the basis of high-resolution co-crystal structures of the cAMP-specific PDE4B and PDE4D with AMP, the cGMP-specific PDE5A with GMP, and the apo-structure of the dual-specific PDE1B. These structures show that an invariant glutamine functions as the key specificity determinant by a "glutamine switch" mechanism for recognizing the purine moiety in cAMP or cGMP. The surrounding residues anchor the glutamine residue in different orientations for cAMP and for cGMP. The PDE1B structure shows that in dual-specific PDEs a key histidine residue may enable the invariant glutamine to toggle between cAMP and cGMP. The structural understanding of nucleotide binding enables the design of new PDE inhibitors that may treat diseases in which cyclic nucleotides play a critical role.     

Cyclic nucleotide phosphodiesterases (PDEs) in human osteoblastic cells; the effect of PDE inhibition on cAMP accumulation

The regulation of the secondary messengers, cyclic adenosine monophosphate (cAMP) and cyclic guanosine monophosphate (cGMP), is crucial in the hormonal regulation of bone metabolism. Both cAMP and cGMP are inactivated by cyclic nucleotide phosphodiesterases (PDEs), a superfamily of enzymes divided into 11 families (PDE1-11). We compared the PDEs of cultured human osteoblasts (NHOst) and SaOS-2 osteosarcoma cells. The PDE activity of NHOst cells consisted of PDE1, PDE3 and PDE7, whereas PDE1, PDE7 and PDE4, but no PDE3 activity was detected in SaOS-2 cells. In line with the difference in the PDE profiles, rolipram, a PDE4 inhibitor, increased the accumulation of cAMP in SaOS-2, but not in NHOst cells. Expression of PDE subtypes PDE1C, PDE3A, PDE4A, PDE4B, PDE7A and PDE7B was detected in both cell types. NHOst cells additionally expressed PDE1A.     

cAMP analogs and selective inhibitors used to study low Km Mucor rouxii cAMP phosphodiesterase.

1. The sensitivity of partially purified low Km phosphodiesterase (PDE) from Mucor rouxii to pharmacological agents and cAMP analogs was studied. The IC50 obtained were compared with those reported for PDEs from higher eukaryotes. 2. The best inhibitors of the hydrolysis of 1 microM cAMP were SQ 65.442 (IC50 c 10 microM), dipyridamol and CI 930. cGMP was not an inhibitor (IC50 greater than 1000 microM). 3. The cAMP analogs were tested as inhibitors of the hydrolysis of 0.1 microM cAMP. 8-Aminohexylamino cAMP was the best inhibitor with an IC50 of c 1 microM. 4. A sedimentation profile of Mucor PDE was assayed in the presence of several pharmacological inhibitors and cAMP analogs. No isoforms with different sensitivity towards the inhibitors were detected. Forms with slightly different behaviour towards some cAMP analogs were observed.     

Phosphodiesterase inhibition by a gastroprotective agent irsogladine: preferential blockade of cAMP hydrolysis

The effect of irsogladine [2,4-diamino-6-(2,5-dichlorophenyl)-s-triazine maleate], an antiulcer drug, on contents of cyclic nucleotides including cAMP and cGMP was investigated in rat stomachs. Irsogladine concentration-dependently increased cAMP content in rat glandula stomach. However, irsogladine at higher concentration (10(-5) M) was unable to further increase cAMP level in the presence of non-selective phosphodiesterase (PDE) inhibitor 3-isobutyl-1-methylxanthine, although 3-isobutyl-1-methylxanthine by itself increased cAMP level. On the other hand, irsogladine had no effect on the glandula cGMP content. Subsequently, the effect of irsogladine on the cyclic nucleotide degradation by purified bovine brain and heart PDEs was investigated. The cAMP degradation by purified bovine brain PDE was partially suppressed by PDE1 inhibitor vinpocetin, PDE2 inhibitor erythro-9-(2-hydroxy-3-nonyl)adenine hydrochloride and PDE4 inhibitor rolipram but not by PDE3 inhibitor cilostamide, and completely inhibited by 3-isobutyl-1-methylxanthine, suggesting that is attributed almost exclusively to PDE1, PDE2 and PDE4. Meanwhile, cGMP degradation by purified bovine brain PDE was partially suppressed by erythro-9-(2-hydroxy-3-nonyl)adenine hydrochloride. Irsogladine preferentially inhibited the response to cAMP degradation compared with cGMP degradation by this brain PDE. The cAMP degradation by bovine heart PDE was almost completely inhibited by the combination with vinpocetine and cilostamide, indicating that is mediated almost exclusively by PDE1 and PDE3. Irsogladine suppressed this cAMP degradation measured in the presence of vinpocetine to almost the same extent as that determined in the presence of cilostamide. These results indicate that irsogladine produces the increase of intracellular cAMP content via non-selective inhibition of PDE isozymes, which may be a key mechanism involved in its gastroprotective actions.     

Alterations in EDHF-type relaxation and phosphodiesterase activity in mesenteric arteries from diabetic rats

In isolated superior mesenteric artery rings from age-matched control rats and streptozotocin (STZ)-induced diabetic rats, we investigated the role of cAMP in endothelium-derived hyperpolarizing factor (EDHF)-type relaxation. The ACh-induced EDHF-type relaxation was significantly weaker in STZ-induced diabetic rats than in control rats, and in both groups of rats it was attenuated by 18alpha-glycyrrhetinic acid (18alpha-GA), an inhibitor of gap junctions, and enhanced by IBMX, a cAMP-phosphodiesterase (PDE) inhibitor. These enhanced EDHF-type responses were very similar in magnitude between diabetic and age-matched control rats. The EDHF-type relaxation was enhanced by cilostamide, a PDE3-selective inhibitor, but not by Ro 20-1724, a PDE4-selective inhibitor. The expression levels of the mRNAs and proteins for two cAMP PDEs (PDE3A, PDE3B) were significantly increased in STZ-induced diabetic rats, but those for PDE4D were not. We conclude that the impairment of EDHF-type relaxations in STZ-induced diabetic rats may be attributed to a reduction in the action of cAMP via increased PDE activity.     

Presence of cyclic nucleotide phosphodiesterases PDE1A,existing as a stable complex with calmodulin,and PDE3A in human spermatozoa

Mammalian sperm motility, capacitation, and the acrosome reaction are regulated by signal transduction systems involving cAMP as a second messenger. Levels of cAMP are controlled by two key enzymes, adenylyl cyclase and phosphodiesterases (PDEs), the latter being involved in cAMP degradation. Calmodulin-dependent PDE (PDE1) and cAMP-specific PDE (PDE4) activities were previously identified in spermatozoa via the use of specific inhibitors. Here we report that human sperm PDEs are associated with the plasma membrane (50%-60%) as well as with the particulate fraction (30%-50%) and have more affinity for cAMP than cGMP. Immunocytochemical data indicated that PDE1A, a variant of PDE1, is localized on the equatorial segment of the sperm head as well as on the mid and principal pieces of the flagellum, and that PDE3A is found on the postacrosomal segment of the sperm head. Immunoblotting confirmed the presence of PDE1A and PDE3A isoforms in spermatozoa. Milrinone, a PDE3 inhibitor, increased intracellular levels of cAMP by about 15% but did not affect sperm functions, possibly because PDE3 represents only a small proportion of the sperm total PDE activity (10% and 25% in Triton X-100 soluble and particulate fractions, respectively). PDE1A activity in whole sperm extract or after partial purification by anion-exchange chromatography was not stimulated by calcium + calmodulin. Results obtained with electrophoresis in native conditions indicated that calmodulin is tightly bound to PDE1A. Incubation with EGTA + EDTA, trifluoperazine, or urea did not dissociate the PDE1A-calmodulin complex. These results suggest that PDE1A is permanently activated in human spermatozoa.