Thesis (Ph. D.)--University of Lund, 1992. / Published dissertation. Includes bibliographical references.

Thesis (Ph. D.)--University of Lund, 1992. / Published dissertation. Includes bibliographical references.

Thesis advisor: Charles S. Hoffman / The biological roles of the most recently discovered mammalian cyclic nucleotide phosphodiesterase (PDE) family, PDE11, are poorly understood, in part due to the lack of selective inhibitors. To address this need for such compounds I completed a ~200,000 compound high throughput screen (HTS) for PDE11 inhibitors using a yeast-based growth assay. Further characterization of lead candidates using both growth-based assays in the fission yeast Schizosaccharomyces pombe and in vitro enzyme assays identified four potent and selective PDE11 inhibitors. I examined the effect of these compounds on human adrenocortical cells, where PDE11 is believed to regulate cortisol levels. One compound, along with two structural analogs, elevates cAMP levels and cortisol production through PDE11 inhibition, thus phenocopying the behavior of adrenocortical tumors associated with Cushing syndrome. These compounds can be used as research tools to study the biological function of PDE11, and can also serve as leads to develop therapeutic compounds for the treatment of adrenal insufficiencies. This study further validates the yeast-based HTS platform as a powerful tool for the discovery of potent, selective and biologically-active PDE inhibitors. / Thesis (PhD) — Boston College, 2012. / Submitted to: Boston College. Graduate School of Arts and Sciences. / Discipline: Biology.

No description available.

Dissertation (Ph.D.)--Auburn University,2009. / Abstract. Vita. Includes bibliographic references.

Pathogenic bacteria such as M.tuberculosis have evolved several mechanisms to aid their intracellular survival and subvert host defenses. One of the contributing factors is thought to be the production and secretion of large amount of cAMP, Mycobacterial genomes encode a large number of adenylyl cyclases distinct in their structure and regulatory mechanisms. The roles of these enzymes in the physiology and pathogenesis of virulent mycobacteria are only now being elucidated. The roles of phosphodiesterases (PDEs), which serve to lower cAMP levels through degradation are, however, relatively unexplored. The Rv0805 gene was previously shown to code for an active phosphodiesterase from Mycobacterium tuberculosis. Bioinformatics analysis revealed that orthologs of Rv0805 were found even in eukaryotes. Biochemical and structural characterization of Rv0805 revealed that it was a class III cAMP phosphodiesterase. Comparative genomics identified a close ortholog of Rv0805 in M. leprae (ML2210). The genome of M. leprae Encodes only 1,604 predicted proteins and possesses the highest number of pseudogenes, 1,116. The retention of a functional PDE, the ortholog of Rv0805, in the minimal genome of M. leprae is indicative of its importance in cellular physiology. Biochemical characterization of proteins from M. leprae and use of heterologous hosts will help understand this human pathogen better, since there are no tools currently available to genetically manipulate this bacterium. Sequence analysis of ML2210 revealed the presence of conserved motifs and residues known to be critical for catalysis and unique to class III phosphodiesterases. ML2210 shares 83% sequence identity with Rv0805 and 24% sequence identity with the phosphodiesterase from E. coli (cpdA). In vitro biochemical characterization of ML2210 using non-nucleotide colorigenic and cyclic nucleotide substrates revealed that it was an enzymatically active phosphodiesterase. Kinetic parameters of ML2210 with respect ot colorigenic substrates revealed that its catalytic properties were similar to that of Rv0805. However, with respect to hydrolysis of 3’, 5’-cAMP, ML2210 was catalytically more efficient than Rv0805, suggesting that in spite of being orthologs, these enzymes have evolved distinct specificities at their active site. A parallel of monoclonal antibodies raised to Rv0805 was also used understand the differences in the biochemical properties of Rv0805 and ML2210 better. It was observed that only one monoclonal antibody was able to recognize ML2210 by ELISA and not by Western blot analysis. These results revealed that conformational differences between ML2210 and Rv0805 exist. Over-expression of ML2210 in M. smegmatis resulted in a modest decrease in intracellular cAMP levels. Despite the absence of a predicted transmembrane region or a membrane-targeting signal, ML2210 localized to the cell envelop fraction upon over expression in M. smegmatis. Moreover, like Rv0805, over-expression of ML2210 also resulted in perturbation of the cell wall of M. smegmatis, arguing for additional cellular roles of this protein. Orthologs of Rv0805 or ML2210 are found only in slow growing mycobacteria suggesting that other cyclic nucleotide phosphodiesterases could regulate cAMP levels in fast growing mycobacteria like M. smegmatis. Since BLAST results did not retrieve an ortholog of Rv0805 or ML2210, COG1409 (COG database) containing Rv0805 was examined for the presence of other mycobacterial phosphodiesterases. Bioinformatics analysis identified Rv2795c as another PDE from M. tuberculosis. Sequence analysis of Rv2795c revealed the presence of all the motifs conserved in the class III PDEs but Rv2795c shared only 22% sequence identity with Rv0805 and 19% sequence identity with CpdA. Importantly, an ortholog of Rv2795c was identified in M. leprae. Interestingly. Rv2795c and its orthologs branched away from Rv0805, making it phylogenetically distinct and hence warranting further characterization. Recombinant, purified MSMEG_2647 (the Rv2795c ortholog from M. smegmatis) was able to hydrolyze cyclic nucleotides and other phosphodiester substrates in vitro. The Km for colorigenic substrates was higher when compared to the Km of ML2210 or Rv0805 for these substrates. However, the kinetic parameters of MSMEG_2647 for cyclic nucleotides were comparable to those of ML2210 or Rv0805. MSMEG_2647 was a metal dependent enzyme and among the panel of metals tested, Mn2+ supported the highest in vitro catalytic activity of MSMEG_2647. Zn2+ inhibited the catalytic activity of MSMEG_2647. In order to gain insight into the catalysis of MSMEG_2647, the end products of cAMP hydrolysis by MSMEG_2647 were analysed using reverse phase HPLC. The assay revealed that the end products of cyclic nucleotide hydrolysis by MSMEG_2647 were different when compared to the end products of hydrolysis of the same substrates by Rv0805 or ML2210. This suggests differences in the architecture of the active site residues of the mycobacterial MPEs. A mutational anlaysis of the active site residues in MSMEG_2647 was carried out to identify residues involved in substrate recognition and metal coordination. Although Rv0805 and MSMEG_2647 shared only a 22% sequence identity, MSMEG_2647 displayed strict conservation in the core MPE motifs. Mutation of the active residues N97 and H98 in Rv0805 had led to an abrogation of its catalytic activity. However, corresponding mutations of N76A and H77A in MSMEG_2647, did not lead to a loss in its catalytic activity. A third mutation known to be important for the catalytic activity of Rv0805 (D19) was incorporated. The corresponding residue at D19 position was mutated to an alanine. The catalytic activity of MSMEG_2647D19AN76AH77A mutant was abrogated, suggesting that while the core MPE motifs are conserved between mycobacterial PDEs, differences in the ensemble of the active site residues contributing to their catalytic activity exist. Thus, at least two biochemically diverse PDE clades are found in mycobacterial species. In order to decipher the function of MSMEG_2647, its expression was monitored during the growth of M. Smegmatis. The promoter of MSMEG_2647 displayed maximum activity during the logarithmic phase of M. smegmatis growth after which its activity declined as M. smegmatis entered the stationary phase. However in contrast to this, the transcript corresponding to msmeg_2647 mRNA was found at both logarithmic and stationary phases. The MSMEG_2647 protein was also detected at both logarithmic and stationary phases of M. smegmatis. These results suggest that additional factors may contribute to the stability of msmeg_2647 mRNA and protein levels. Localization studies of MSMEG_2647 revealed that MSMEG_2647 was present in the cytosol as well as in the cell envelope fractions. Interestingly, over-expression of MSMEG_2647 did not result in a significant increase in PDE activity in various subcellular fractions, suggesting tight regulation on the in vivo activity in various subcellular fractions, suggesting tight regulation on the in vivo activity of MSMEG_2647. In addition, over-expression of MSMEG_2647 in M. smegmatis led to only a modest decrease in cAMP levels in M. smegmatis. These results suggested additional roles of MSMEG_2647 in the biology of mycobacteria. Overexpression of MSMEG_2647 peturbed the integrity of cell wall as assessed by the use of lipophillic indicators of cell growth, crystal violet and malachite green, and a cell wall targeting antibiotic, isoniazid. Analyzing the gene neighborhood of MSMEG_2647 provided an insight into its putative function. It was observed that the stop codon of msmeg_2647 overlapped with the start codon of msmeg_2648 and stop codon of msmeg-2648 overlapped with the start codon of msmeg_2649. RT PCR was carried out at logarhtimic and stationary phases of M. smegmatis growth, which revealed that a polycistronic mRNA was being transcribed. These results confirmed that msmeg_2647, msmeg_2648 and msmeg_2649 were a part of an operon. Interestingly, these three genes as a gene cluster were confined to only those actinobacteria that produced mycolic acids. An operon often encodes products that form multiprotein complexes and operate in a common pathway. Since there were a part of an operon, a GST pull-down approach was employed to test if MSMEG_2647, MSMEG_2648 and MSMEG_2649 could physically interact. It was observed that MSMEG_2647 interacted with MSMEG_2648 and MSMEG_2649. MSMEG_2648 in turn interacted with MSMEG_2649. A role for MSMEG_2647 as a scaffold recruiting MSMEG_2648 and MSMEG_2649 is therefore proposed. In turn, a complex formation with these proteins may regulate the activity of MSMEG_2647. Attempts to generate a knock out of msmeg_2647 in M. smegmatis by homologous recombination were not successful suggesting either the gene was essential or a polar effect on msmeg_2648(an essential gene for the viability of M. smegmatis) may not allow msmeg_2647 to be deleted from the genome of M. smegmatis. In summary, this study has identified and characterized two new phosphodiesterases from mycobacteria, one from the pathogenic mycobacterium, M. leprae and the other, a PDE from M. smegmatis that is conserved in all species of mycobacteria. Several, key biochemical differences were observed using biochemical and biological approaches. It appears that the cellular roles of mycobacterial phsophodiesterases may extend beyond cAMP hydrolysis, with these proteins not only regulating cell wall properties but also acting as scaffolding proteins in the cell.

O óxido nítrico (NO) é um mensageiro químico encontrado em diversos tipos celulares como células endoteliais, neurônios e macrófagos. A síntese do NO é realizada pela ação da enzima óxido nítrico sintase (NOS). Um dos mecanismos de ação do NO é dado pela ativação da enzima guanilato ciclase solúvel (GCs), resultando na produção de monofosfato cíclico de guanosina (GMPc), um mensageiro secundário nessa via de sinalização celular. O GMPc por sua vez é capaz de modular a atividade de algumas fosfodiesterases (PDEs), enzimas responsáveis pela degradação do GMPc e de outro nucleotídeo cíclico, o monofosfato cíclico de adenosina (AMPc). O objetivo deste trabalho foi investigar os efeitos da elevação dos níveis de NO por meio do doador de óxido nítrico (SNAP) e o uso de inibidores de diferentes isoformas de fosfodiesterases no meio de cultivo durante a maturação in vitro (MIV) de oócitos bovinos sobre a retomada da meiose, concentração de NO e níveis de GMPc e AMPc. Deste modo, os complexos cumulus-oócito (CCOs) bovinos foram cultivados por até 9 horas com o doador de NO (SNAP - 10-7 M) associado ou não ao inibidor de GCs (ODQ - 10-5 M) e associado ou não aos inibidores das fosfodiesterases, PDE5 (Sildenafil - 10µM), PDE3 (Cilostamide - 20µM) e PDE8 (Dipiridamole - 50µM). As amostras foram avaliadas quanto a taxa de retomada da meiose, níveis de NO (9h de MIV) e níveis dos nucleotídeos cíclicos GMPc e AMPc (0, 1, 2 e 3h de MIV). O SNAP retardou o rompimento da vesícula germinativa com 9horas de cultivo (P<0,05) e quando o SNAP foi associado ao ODQ o efeito foi revertido (P>0,05). A inclusão de SNAP no cultivo, os níveis de NO foram elevados (P<0,05). Os níveis de GMPc só foram influenciados positivamente pelo SNAP com 1 hora de cultivo (P<0,05) e após 2 e 3 horas, esta influência não persistiu (P>0,05), visto que o ODQ aboliu o efeito. A influência do SNAP foi devido ao estímulo da GCs. Para os níveis de AMPc, o doador de NO não foi capaz de influenciar suas concentrações durante as 3 horas de cultivo (P>0,05). Quando o SNAP foi associado ao Sildenafil (SNAP+SIL) não houve diferença em relação ao grupo imaturo (P>0,05), porém, também não se diferiu do tratamento SNAP e controle (P>0,05). Para as taxas de retomada de meiose todos os tratamentos foram eficientes e conseguiram retardar a quebra da vesícula germinativa diante do grupo controle (P<0,05), sendo o grupo SNAP+CIL mais eficiente. Para os níveis de AMPc, nem mesmo com a utilização de inibidores das PDE3 e PDE8 foi possível atenuar a queda do nucleotídeo. Em conclusão, o SNAP exerceu influência na retomada da meiose, na concentração de NO e nos níveis de GMPc sendo que sua ação se deve à atividade da GCs. Não houve influência sobre os níveis de AMPc, quando o SNAP foi associado a inibidores específicos de fosfodiesterases, mesmo quando apresentaram efeito sobre a retomada da meiose. A via NO/GCs/GMPc não parecer atuar sobre a via PDE3/AMPc, sugerindo a ação de outras vias no controle da meiose. / Nitric oxide (NO) is a chemical messenger found in many cell types such as endothelial cells, neurons and macrophages. The synthesis of NO is made by the action of nitric oxide synthase (NOS). One of the mechanisms of action of NO is given by the activation of the enzyme soluble guanylate cyclase (sGC), resulting in the production of cyclic guanosine monophosphate (cGMP), a secondary messenger in this pathway of cell signaling. The cGMP in turn is capable of modulating the activity of some phosphodiesterases (PDEs), enzymes responsible for degradation of cGMP and other cyclic nucleotide, cyclic adenosine monophosphate (cAMP). The objective of this study was to investigate the effects of elevated levels of NO via nitric oxide donor (SNAP) and the use of inhibitors of phosphodiesterase isoforms in the culture medium during in vitro maturation (IVM) of bovine oocytes on resumption of meiosis, the concentration of NO and cGMP and cAMP levels. Thus, the complexes cumulus-oocyte (COCs) were cultured for cattle up to 9 hours with the NO donor (SNAP - 10-7 M) with or without the inhibitor of sGC (ODQ - 10-5 M) and with or without to inhibitors of phosphodiesterase PDE5 (Sildenafil - 10µM), PDE3 (Cilostamide - 20µM) and PDE8 (Dipyridamole - 50µM). The samples were evaluated for the rate of resumption of meiosis, levels of NO (9h - IVM) and levels of cyclic nucleotides cGMP and cAMP (0, 1, 2 and 3h - IVM). The SNAP delayed with germinal vesicle breakdown of 9hours cultivation (P < 0.05) when SNAP was associated with ODQ was reversed the effect (P> 0.05). The inclusion of SNAP cultivation, NO levels were increased (P<0.05). cGMP levels were positively influenced only by the snap 1 hour in culture (P<0.05) and after 2 and 3 hours, this effect was not maintained (P<0.05), whereas ODQ abolished the effect. The influence of SNAP was due to stimulation of GCs. For cAMP levels, the NO donor was not able to influence their concentrations during the 3 h incubation (P>0.05). When SNAP was associated with Sildenafil (SIL+SNAP) there was no difference compared to the immature group (P>0.05), however, also did not differ from SNAP treatment and control (P>0.05). Rates for resumption of meiosis all treatments were efficient and able delay before germinal vesicle breakdown in the control group (P<0.05), with SNAP+CIL group more efficient. For cAMP levels, even with the use of inhibitors of PDE3 and PDE8 was possible to alleviate the decrease of the nucleotide. In conclusion, SNAP exerted influence on the resumption of meiosis, the concentration of NO and cGMP levels and that its action is due to a GCs activity. There was no effect on cAMP levels, when SNAP was associated with specific phosphodiesterase inhibitors, even when presented effect on the resumption of meiosis. The pathway NO/sGC/cGMP seem not act on the pathway PDE3/AMPc, suggesting the action of other pathways in the control of meiosis.

La protéine kinase AMPc-dépendante (PKA) joue un rôle crucial dans la régulation neurohormonale de la fonction cardiaque. L’activation aiguë de la PKA est bénéfique car elle conduit à une augmentation de la contraction cardiaque en phosphorylant les acteurs clés du couplage excitation-contraction. En revanche, son activation chronique est délétère et ces effets semblent faire intervenir la régulation de protéines nucléaires pouvant conduire au remodelage hypertrophique et à l'insuffisance cardiaque. La localisation subcellulaire de la PKA, assurée par des protéines d’ancrage (AKAPs), est importante pour la rapidité et la spécificité d’action des hormones mettant en jeu la voie de l’AMPc. Les niveaux d’AMPc sont régulés par l’activité des adénylate cyclases et des phosphodiestérases (PDEs), et l’état de phosphorylation des protéines cibles de la PKA dépend de l’activité des Ser/Thr phosphatases (PPs). Dans le cœur, les PDEs les plus importantes dégradant l’AMPc sont les PDE3 et les PDE4. Les principales PPs cardiaques sont PP1, PP2A et PP2B. Dans une première partie de mon travail, j’ai mis au point, dans les cardiomyocytes de rats adultes, une mesure de l’activité de la PKA en temps réel dans les compartiments cytoplasmiques et nucléaires. J’ai utilisé pour cela des sondes de type AKAR (A-kinase activity reporters) basées sur le transfert d’énergie de fluorescence (FRET) et localisées spécifiquement dans le noyau ou dans le cytoplasme par des séquences d’adressage ou d’exclusion nucléaires. J’ai ainsi pu montrer qu’une stimulation maintenue des récepteurs β-adrénergiques active la PKA de façon plus importante dans le cytoplasme que dans le noyau, et que cette activation se développe lentement au niveau nucléaire que dans le cytoplasme. De ce fait, une stimulation brève des récepteurs β-adrénergiques active maximalement la PKA dans le cytoplasme, mais de façon marginale dans le noyau. Dans une seconde partie de l’étude, je me suis intéressée au rôle des PDE3 et PDE4 ainsi qu’à celui de PP1, PP2A et PP2B dans la régulation de l’activité PKA cytoplasmique et nucléaire, en réponse à une stimulation β-adrénergique. J’ai montré que la PDE4, mais pas la PDE3, régule l’activité de la PKA cytoplasmique et nucléaire. L’utilisation de souris invalidées pour les gènes Pde4b et Pde4d a révélé que l’isoforme PDE4B est prédominante pour la modulation de l’activité PKA cytoplasmique, alors que les deux isoformes PDE4B et PDE4D contribuent à la régulation de l’activité PKA nucléaire. Finalement, j’ai montré que la PP1 et la PP2A, mais pas la PP2B, participent à la terminaison des réponses β-adrénergiques dans le cytoplasme, alors qu’au niveau nucléaire, la PP1 semble jouer un rôle majeur. En conclusion, ce travail a mis en évidence le rôle des phosphodiestérases et des phosphatases dans l’intégration différentielle des réponses PKA à une stimulation β-adrénergique dans le cytoplasme et le noyau de cardiomyocytes adultes. / The cAMP-dependent protein kinase (PKA) exerts short term beneficial effects on cardiac function by phosphorylating several key excitation-contraction coupling (ECC) proteins. However, its chronic activation is deleterious on the long term, and this may involve regulation of nuclear effectors ultimately leading to hypertrophic remodelling and heart failure. The subcellular localization of PKA, mediated by anchoring proteins (AKAPs), is important for the speed and specificity of hormones that activate the cAMP pathway. The levels of cAMP are regulated by adenylyl cyclase and phosphodiesterases (PDEs), and PKA activity is counterbalanced by Ser/Thr phosphatases (PPs). In heart, the most important PDEs that degrade cAMP belong to the PDE3 and PDE4 famillies, whereas the major cardiac PPs are PP1, PP2A and PP2B. In a first part, I developed, in adult rat cardiomyocytes, a technique to measure PKA activity in real time specifically in the cytoplasm and the nucleus. For this I used genetically-encoded fluorescence resonance energy transfer (FRET) sensors called AKAR (A-kinase activity reporters) that can be targeted specifically to the nucleus or the cytoplasm by nuclear localization or exclusion sequences, respectively. Using this approach, I showed that maintained β-adrenergic stimulation activates PKA more efficiently and more potently in the cytoplasm than in the nucleus, and that the kinetics of PKA activation was much slower in the nucleus than in the cytoplasm. Accordingly, a short β-adrenergic stimulation maximally activated PKA in the cytoplasm but marginally activated PKA in the nucleus. In a second part, I characterized the respective contribution of PDE3, PDE4, and PP1, PP2A and PP2B families in the regulation of cytoplasmic and nuclear PKA activity in response to β-adrenergic stimulation. PDE4, but not PDE3, regulates PKA activity in the cytoplasm and in the nucleus. The use of knock out mice for Pde4b and Pde4d genes revealed that PDE4B plays a predominant role to modulate β-AR stimulation of cytoplasmic PKA, whereas in the nucleus both PDE4B and PDE4D isoforms contribute. Finally, I showed that both PP1 and PP2A, but not PP2B, participate to the termination of β-adrenergic PKA responses in the cytoplasm, whereas PP1 appears to play a major role in the nuclei. In conclusion, this work highlights the role of phosphodiesterases and phosphatases in the differential integration of PKA responses to β-adrenergic stimulation in the cytoplasm and the nucleus of adult cardiomyocytes.

L’objectif de mon travail de thèse était d’une part, de mieux comprendre le rôle des différentes familles de phosphodiestérases (PDEs) dans la régulation de la signalisation dépendante de l’AMPc (PDE-AMPc) dans les cellules musculaires lisses vasculaires (CMLVs), et d’autre part, d’évaluer leur implication fonctionnelle dans la réactivité vasculaire et leur altération potentielle dans un modèle physiopathologique, l’insuffisance cardiaque (IC). Mon travail s’est articulé autour de deux modèles de muscle lisse vasculaire : (1) des CMLVs isolées en culture ayant acquis un phénotype synthétique sur lesquelles une approche d’imagerie en temps réel (FRET : Transfert d’Energie de Fluorescence par Résonance) a été appliquée afin de visualiser in situ la dynamique spatiotemporelle des signaux dépendants de l’AMPc. Nos résultats indiquent que, dans ces cellules, l’augmentation des niveaux d’AMPc provoquée par la stimulation β-adrénergique (β-AR) implique différents récepteurs suivant le compartiment intracellulaire considéré (β1- et β2-ARs dans le cytosol et seulement β2-ARs dans le compartiment sous-membranaire). Nous avons par ailleurs observé que l’expression des ARNm des différentes isoformes de PDE-AMPc et la contribution fonctionnelle de ces enzymes dans la régulation des signaux AMPc intracellulaires étaient dépendantes de la densité des CMLVs en culture.(2) des anneaux d’artères intactes issues de deux lits vasculaires différents (aorte et artère mésentérique) isolées à partir de rats sains et IC, permettant d’étudier leur fonction contractile et donc la régulation de celle-ci par la voie de l’AMPc. Nous avons montré que les familles de PDE-AMPc contribuent de façon différente au contrôle du tonus vasculaire dans l’aorte thoracique (PDE3 = PDE4 sans participation de la PDE2) et dans l’artère mésentérique (PDE4 > PDE2 sans participation de la PDE3), l’endothélium exerçant un rôle essentiel dans la régulation de l’activité de ces PDEs musculaires lisses, notamment par le biais de la production de NO. Nous avons également mis en évidence des altérations de la réactivité vasculaire, et notamment de son contrôle par la voie de l’AMPc/PDE, dans notre modèle de rat IC. Dans l’aorte, la dysfonction endothéliale liée à l’altération de la voie du NO est à l’origine d’une augmentation de l’activité PDE3 masquant l’activité PDE4 et la relaxation β-adrénergique. Dans l’artère mésentérique des rats IC, dont la fonction endothéliale apparaît préservée, les PDE2, 3 et 4 restent fonctionnelles.L’ensemble de nos travaux souligne le rôle essentiel des PDEs dans la régulation de la signalisation AMPc vasculaire, et montre que l’activité et la fonction des différentes familles de PDE-AMPc sont finement modulées par de nombreux paramètres (phénotype et densité cellulaire des CMLVs) ou situations physio-pathologiques (nature du lit vasculaire, présence de l’endothélium, situation d’IC). / The aim of my thesis was to investigate the role of cyclic nucleotide phosphodiesterases (cAMP-PDEs) in the regulation of cAMP-dependent signaling in vascular smooth muscle cells (VSMCs), and to assess their functional involvement in vascular reactivity and their potential alteration in a pathophysiological model of heart failure (HF). My work was based on two models of vascular smooth muscle:(1) Isolated VSMCs in culture having acquired a synthetic phenotype, in which an approach of real-time imaging (FRET: Fluorescence Resonance Energy Transfer) was applied in situ to visualize the spatiotemporal dynamics of cAMP-dependent signals. Our results indicate that, in these cells, increased levels of cAMP induced by β-adrenergic stimulation (β-AR) involve different β-ARs subtypes according to the intracellular compartment considered (β1-and β2-ARs in the cytosol and only β2-ARs in the submembrane compartment). We also observed that the mRNA expression of cAMP-PDEs isoforms and the functional contribution of these enzymes in the regulation of intracellular cAMP signals were dependent on the VSMCs seeding density in culture.(2) Arterial blood vessels from two different vascular beds (aorta and mesenteric artery) isolated from healthy and HF rats, to study their contractile function and thus the regulation by the cAMP pathway. We showed that cAMP-PDE families contribute differently to the control of vascular tone in the thoracic aorta (PDE3 = PDE4, no PDE2) and mesenteric artery (PDE4 > PDE2, no PDE3), endothelium exerting a crucial role in the regulation of their functional activities, especially through the production of nitric oxide (NO). We also demonstrated alterations in vascular reactivity during HF, including its control through the cAMP-PDEs. In the aorta, endothelial dysfunction associated with the alteration of the NO pathway leads to an increase in PDE3 activity which masks PDE4 activity and β-AR relaxation. In mesenteric artery from HF rats, endothelial function is preserved and PDE2, 3 and 4 are functional.This study underlines the importance of PDEs in regulating vascular cAMP signaling, and shows that the activity and function of different cAMP-PDE families are tightly modulated by many parameters (VSMCs phenotype and seeding density) and/or physiopathological situations (vascular bed, endothelium and HF).

La voie β-adrénergique (β-AR)/AMPc est cruciale pour l’adaptation de la fonction cardiaque. Dans l’insuffisance cardiaque (IC), cette signalisation est perturbée et une part importante des patients meurt de troubles du rythme. Classiquement, les effets inotrope et lusitrope positifs de l’AMPc sont attribués à la phosphorylation par la protéine kinase AMPc dépendante (PKA) des protéines clés du couplage excitation–contraction (CEC). L’AMPc active aussi le facteur d’échange Epac, impliqué dans l’hypertrophie cardiaque et le contrôle de l’homéostasie calcique. Une cible d’Epac est la CaMKII, une kinase modulée par le Ca2+ et la calmoduline qui phosphoryle aussi les protéines clés du CEC, et dont l’activation est pro-arythmique.Les phosphodiestérases (PDEs) de type 3 et 4 sont majeures pour dégrader l’AMPc et contrôler l’homéostasie calcique et le CEC. Les inhibiteurs de PDE3 sont de puissants cardiotoniques mais leur utilisation est limitée par leurs effets pro-arythmiques. De plus, l’invalidation de gènes codant pour PDE4 conduit à des arythmies ventriculaires. Mon travail a permis d’identifier les perturbations de l’homéostasie calcique responsables de la survenue d’arythmies lorsque l’activité des PDE3 et des PDE4 est diminuée. Mes résultats montrent que les inhibiteurs de PDEs exercent des effets inotropes via PKA, mais suscitent des vagues de Ca2+ pro-arythmiques impliquant la PKA et la CaMKII activée en partie via Epac. Ceci suggère l'utilisation potentielle d'inhibiteurs de CaMKII comme compléments aux inhibiteurs de PDEs pour limiter leurs effets délétères, une hypothèse que j’ai pu vérifier dans un modèle porcin plus proche du patient. / The β-adrenergic pathway (β-AR)/cAMP is crucial for the adaptation of the cardiac function upon stress. In heart failure (HF), this signaling pathway is disrupted and a significant proportion of patients dies of cardiac arrhythmias. Classically, the inotropic and lusitropic effects of cAMP are attributed to the phosphorylation by the cAMP-dependent protein kinase (PKA) of the key proteins of the excitation-contraction coupling (ECC). cAMP also activates the exchange factor Epac, which is involved in cardiac hypertrophy and controls intracellular Ca2+ homeostasis. Epac activates CaMKII, another kinase modulated by Ca2+ and calmodulin which phosphorylates the same key proteins of the ECC, and is involved in arrhythmogenesis.Phosphodiesterases (PDEs) type 3 and 4 are crucial enzyme to degrade cAMP and to control Ca2+ homeostasis, thus ECC. PDE3 inhibitors are potent cardiotonic drugs but their use is limited by their pro-arrhythmic effects. Furthermore, the invalidation of genes encoding PDE4 results in ventricular arrhythmias. My work allowed characterizing the perturbations of Ca2+ homeostasis which lead to arrhythmias when PDE3 and PDE4 activities are decreased. My results show that PDE inhibitors exert inotropic effects via PKA, but evoke pro-arrhythmic Ca2+ waves via both PKA and CaMKII, the latter being activated in part via Epac. Altogether, these results suggest the potential use of CaMKII inhibitors as adjuncts to PDEs inhibitors to limit their deleterious effects, a hypothesis I also tested in a porcine model closer to the patient.