The guanosine monophosphate synthetase (GMPS) is a class I glutamine amidotransferase, involved in the de-novo purine nucleotide biosynthesis. The enzyme catalyzes the biochemical transformation of xantosine (XMP) into guanosine monophosphate (GMP) in presence of ATP, Mg2+ and glutamine. All GMPSs consist of two catalytic sites 1) for GATase activity 2) for the ATPPase activity. The two catalytic sites may be housed in the same polypeptide (two-domain-type) or in separate polypeptides (two-subunit-type). Most of the studies have been performed on two-domain-type GMPSs, while only one study has been reported from two-subunit-type GMPS (Maruoka et al. 2009). The two-subunit-type GMPS presents an example where the component reactions of a single enzymatic reaction are carried out by two distinct subunits. In order to get better understanding of structural aspects and mechanistic principle that governs the GMPS activity in two-subunit-type GMPSs, we initiated the study by taking GMPS of Methanocaldococcus jannaschii as a model system. The GMPS of M. jannaschii (Mj) is a two-subunit-type protein. The GATase subunit catalyzes the hydrolysis of glutamine to produce glutamate and ammonia. The ATPPase subunit catalyses the amination of XMP to produce GMP using the ammonia generated in GATase subunit. Since the two component reactions are catalysed by two separate subunits and are coupled in the way that product of one reaction (ammonia) acts as a nucleophile in the second reaction. The cross-talk between these two subunits in order to maximise the efficiency of overall GMPS warrants investigation. The GATase activity is tightly regulated by the interaction with ATPPase domain/subunit, in all GMPS except in the case of P. falciparum. This interaction is facilitated by substrate binding to the ATPPase domain/subunit. Though, the conditions for the interaction between two subunits is known in a two-subunit-type GMP synthetase from P. horikoshii, the structural basis of substrate dependent interaction is not known. As a first step to understand the structural basis of interaction between the Mj GATase and Mj ATPPase subunits, we have determined the structure of Mj GATase (21 kDa) subunit using high resolution, multinuclear, multidimensional NMR spectroscopy. Sequence specific resonance assignments were obtained through analysis of various 2D and 3D hetero-nuclear multidimensional NMR experiments. NMR based distance restraints were obtained from assignment of correlations observed in NOE based experiments. Data were acquired on isotopically enriched samples of Mj GATase. The structure of Mj GATase (2lxn) was solved by using cyana-3.0 using NMR based restraints as input for the structure calculation. The ensemble of 20 lowest-energy structures showed root-mean-square deviations of 0.35±0.06 Å for backbone atoms and 0.8±0.06 Å for all heavy atoms. Attempts were also made to obtain assignments for the 69.6 kDa dimeric ATPPase subunit. Partial assignments have been obtained for this subunit. The GATase subunit is catalytically inactive. So far, there has been only one published report on a two-subunit-type GMPS from P. horikashii. The study has shown that the catalytic activity of GATase is regulated by the GATase-ATPPase interaction which is facilitated by the substrate binding to the ATPPase subunit. For the first time, we have provided the structural basis of interaction between GATase-ATPPase (112 kDa) in a two-subunit-type GMPS. Observed line width changes were used to identify residues in GATase residues that are involved in the Mj GATase-ATPPase interaction. Our data provides a possible explanation for conformational changes observed in the Mj GATase subunit upon GATase-ATPPase interaction that lead to GATase activation. Ammonia is generated in GATase subunit and is very reactive and labile. Thus, the faithful transportation of ammonia from GATase to ATPPase subunit is very crucial for optimal GMPS activity. Till date, a PDB query for GMPS retrieves only one structure which belongs to two-subunit-type GMPS, where authors have determined the structures of GATase and ATPPase subunits separately. However, the structure of holo-GMPS is not determined yet. Using interface information from experimental data and HADDOCK, we have constructed a model for the holo-GMPS from M. jannaschii. A possible ammonia channel has been deduced using the programs MOLE 2.0 and CAVER 2.0. This ammonia channel has a length of 46 Å, which is well within the range of the lengths calculated for similar channels in other glutamine amidotransferase. It had been suggested earlier that in addition to the magnesium required for charge stabilization of ATP, additional binding sites were present on GMPS. The effect of excess Mg2+ requirement on the GMPS activity has been studied in two-domain-type GMPS. However, the interaction between GATase and Mg2+ has been not investigated in any GMPS. This prompted us to investigate the effect of MgCl2 on Mj GATase subunit. For the first time, using chemical shift perturbation, we have established interaction between Mj GATase and Mg2+. The dissociation constant (Kd) of the Mj GATase-Mg2+ interaction was determined. The Kd value was found to be 1 mM, which indicates a very weak interaction. The substrate of the GATase subunit is glutamine. The condition of the hydrolysis of the glutamine is known in GMPS. However, the binding of the glutamine and associated conformational changes in GATase have been not studied in GMPS. Furthermore, till date there is no structure available for the glutamine bound GMPS/GATase. Using isotope edited one dimensional and two-dimensional NMR spectroscopy; we have shown that the Mj GATase catalytic residues are not in a compatible conformation to bind with glutamine. Thus, a conformational change in Mj GATase subunit is a pre-requisite condition for the binding of glutamine. These conformational changes are brought by the Mj GATase-ATPPase interaction.

Thèse numérisée par la Direction des bibliothèques de l'Université de Montréal. / L'anion superoxyde peut agir comme une molécule de signalisation ou comme un facteur préjudiciable selon sa concentration, l'organe cible, et selon la présence ou non d'antioxydants neutralisants. Actuellement, dans les cellules musculaires lisses (CMLs) vasculaires, les effets de l'anion superoxyde sur les différentes voies de transduction du signal et sur les interactions croisées entre ces voies ne sont pas encore définies. Par conséquent, une meilleure connaissance des effets de l'anion superoxyde sur les différentes voies de signalisation pourrait fournir une meilleure compréhension des mécanismes sous-jacents aux fonctions altérées des CMLs vasculaires observées dans des conditions pathologiques. L'objectif général de cette étude était de caractériser et d'évaluer le rôle modulateur de l'anion superoxyde, produit par la réaction de l'hypoxanthine avec la xanthine oxidase, sur les activités de différentes voies de signalisation dans les CMLs vasculaires, et de déterminer si la sensibilité de différentes voies de signalisation à l'anion superoxyde était altérée dans l'hypertension artérielle. Le projet de ce programme de recherche était basé sur les principaux postulats suivants : (1) l'anion superoxyde pourrait affecter sélectivement la production d'inositol 1,4,5-triphosphates (IP3), de GMPc, ou d'AMPc dans les CMLs vasculaires; (2) le rôle modulateur de l'anion superoxyde pourrait être dû à une altération des interactions croisées entre différentes voies de signalisation; et (3) les anomalies observées dans les CMLs vasculaires chez le rat spontanément hypertendu (SHR) pourraient être reliées à des altérations des différentes voies de signalisation induites par l'anion superoxyde. Une production augmentée d'1P3induite par l'anion superoxyde dans les CMLs d'aorte de rat ou d'artère mésentérique en culture a été démontrée pour la première fois dans cette étude. L'anion superoxyde a augmenté la formation d'IP3d'une manière concentration-dépendante et temps-dépendante. La superoxyde dismutase (SOD), mais non la catalase, a inhibé significativement la formation d'IP3 induite par l'anion superoxyde. L'inhibition de la phospholipase C (PLC) a aboli l'effet de l'anion superoxyde sur la formation d'1P3. La génistéine et la tyrphostine A25, deux inhibiteurs de la tyrosine kinase, ont aussi inhibé significativement la formation d'IP3induite par l'anion superoxyde. L'utilisation d'anticorps anti-PLCy a atténué significativement la formation d'1P3induite par l'anion superoxyde. De plus, le taux d'expression des protéines de la PLCy a été augmenté après l'exposition des CMLs à l'anion superoxyde. Ces observations suggèrent donc que dans les CMLs vasculaires la formation d'1P3 induite par l'anion superoxyde pourrait être en grande partie secondaire à une augmentation de l'activité de la tyrosine kinase liée aux voies de signalisation de la PLCy. En ce qui concerne la voie du GMPc, l'anion superoxyde a diminué significativement les niveaux de base de GMPc et supprimé aussi l'augmentation des niveaux de GMPc induite par des stimulateurs de la guanylyl cyclase, le nitroprussiate de sodium (NPS) ou la s-nitroso-nacétylpénicillamine (SNAP). La formation d'1P3stimulée par l'anion superoxyde a été significativement inhibée par le NPS ou la SNAP, mais potentialisée de façon importante par un inhibiteur de la guanylyl cyclase l'ODQ ou par le KT5823 (un inhibiteur de la protéine kinase dépendant du GMPc). Cependant, l'anion superoxyde n'a pas eu d'effet sur les niveaux de base d'AMPc ou sur la production d'AMPc induite par la forskoline et de plus, l'inhibition de l'adénylyl cyclase ou de la protéine kinase dépendante de l'AMPe n'a pas affecté la formation d'lP3stimulée par l'anion superoxyde. Ces données, par conséquent, suggèrent que l'inhibition de la formation de GMPc par l'anion superoxyde contribue probablement à l'activation de la formation d'1P3induite par l'anion superoxyde en atténuant le rétrocontrôle inhibiteur du GMPc sur les voies de signalisation liées à la PLC, tandis que la voie de signalisation de l'AMPc ne serait pas impliquée dans la formation d'EP3induite par l'anion superoxyde. Dans les CMLs vasculaires de rat SHR, les effets de l'anion superoxyde ont été plus puissants que dans les CMLs de rat WKY, en ce qui concerne l'augmentation de formation d'1P3, la diminution des taux de GMPc et la facilitation induite par l'anion superoxyde des interactions croisées entre les voies du GMPc et de 1'IP3. Dans les CMLs vasculaires des deux souches de rat, la formation d'IP3induite par l'anion superoxyde a été inhibée par une variété d'antioxydants, dont la N-acétylcystéine, l'acide a-lipoïque, la mélatonine et la SOD. Il apparaît donc vraisemblable que l'hypersensibilité à l'anion superoxyde des voies de 1'IP3et du GMPc puissent contribuer à l'augmentation du tonus vasculaire et de la réactivité des CMLs dans l'hypertension artérielle. Nous avons aussi investigué si l'effet de la mélatonine était dû à ses propriétés antioxydantes. Un effet inhibiteur plus important de la mélatonine sur la contraction aortique induite par la norépinéphrine (NE) a été observé chez les rats SHR en comparaison avec les rats Wistar-Kyoto (WKY). L'inhibition par la mélatonine de la formation d'IP induite par la NE a été aussi plus importante dans les CMLs aortiques de rat SHR que dans celles de rat WKY. Les effets plus puissants de la mélatonine chez le rat SHR, qui ont été aussi observés avec la SOD, mais non avec la catalase, ne sont pas dûs à l'activation des récepteurs à la mélatonine ou des récepteurs a-adrénergiques. Ces résultats indiquent que les effets anti-hypertenseurs de la mélatonine sont largement dûs à l'inactivation de l'anion superoxyde, et que les niveaux endogènes d' antioxydants ne parviennent pas à contrecarrer les niveaux accrus d'anion superoxyde produits chez le rat SHR. En conclusion, cette étude révèle une variété de nouveaux mécanismes de signalisation de l'anion superoxyde. Pour la première fois, il a été démontré que l'anion superoxyde active l'hydrolyse des phosphoinositides et augmente les taux d'IP3dans les CMLs vasculaires, principalement par la stimulation de la tyrosine kinase liée à la voie de signalisation de la PLCy. Il a aussi été observé que l'anion superoxyde réduit la formation de GMPc et supprime l'inhibition croisée de 1'1P3par le GMPc, facilitant ainsi la formation d'1P3. Les effets sélectifs de l'anion superoxyde sur les voies de 1'IP3et du GMPc, ainsi que l'existence d'une inhibition croisée de la formation d'1P3par la voie du GMPc, révèlent des mécanismes nouveaux pour expliquer le rôle modulateur de l'anion superoxyde sur les voies de signalisation dans les CMLs. Par conséquent, les effets plus puissants de l'anion superoxyde sur la signalisation de la voie de 1'IP3et de la voie du GMPc dans les CMLs vasculaires de rat SHR, effets qui ont été démontrés pour la première fois dans cette étude, pourraient être responsables des altérations des mécanismes de transduction du signal cellulaire chez le rat SHR et ainsi contribuer au développement et/ou au maintien de l'hypertension artérielle. Ces observations permettent donc d'imaginer de nouvelles orientations pour le développement de nouvelles stratégies pour la prévention ou le traitement de l'hypertension artérielle. / Superoxide anion can act as a signalling molecule or a detrimental factor depending on its concentration, the targeted organ, and the presence of counteracting antioxidants. The effects of superoxide on different signal transduction pathways and on the cross-talk interactions among these pathways in vascular smooth muscle cells (SMCs) are presently still unsettled. Therefore, a better knowledge on the effects of superoxide on different signalling pathways may provide a better understanding of the mechanisms underlying the altered functions in vascular SMCs observed in pathological conditions. The general objective of this study was to characterize and evaluate the modulating role of superoxide generated by the hypoxanthine and xanthine oxidase reaction on the activities of different signalling pathways in vascular SMCs and to investigate whether the sensitivities of different signalling pathways to superoxide were altered in hypertension. The design of the present research program was based on the following major postulates. (1) superoxide might selectively affect the generation of inositol 1,4,5-triphosphates (IP3), cGMP, or cAMP in vascular SMCs; (2) the modulating role of superoxide might be mediated by alteration in the cross-talk interactions among different signalling pathways; and (3) the abnormalities observed in vascular SMCs from spontaneously hypertensive rats (SHR) might be related to the alterations induced by superoxide on different signalling pathways. An enhanced production of 1P3induced by superoxide in cultured SMCs from rat aorta or mesenteric artery was demonstrated, for the first time, in this study. Superoxide increased 1P3 formation in a concentration- and time-dependent manner. Superoxide dismutase (SOD), but not catalase, significantly inhibited the superoxide-increased 1P3formation. The inhibition of phospholipase C (PLC) abolished the effect of superoxide on IP3formation. Genistein and tyrphostin A25, two tyrosine kinase inhibitors, also significantly inhibited the superoxideinduced IP3formation. The application of antibody against PLCI, significantly attenuated the superoxide-induced 1P3formation. Moreover, the expression level of PLC7proteins was increased after exposing SMCs to superoxide. These observations thus suggest that superoxideinduced IP3 formation may be in a great part secondary to an increase in the activity of tyrosine kinase-link PLCy signalling pathways in vascular SMCs. Concerning the cGMP pathway, superoxide significantly decreased the basal levels of cGMP and also suppressed the rise in cGMP levels induced by guanylyl cyclase stimulator sodium nitroprusside (SNP) or s-nitroso-n-acetylpenicillamine (SNAP). The superoxide-induced IP3 formation was significantly inhibited by SNP or SNAP, but markedly potentiated by a guanylyl cyclase inhibitor ODQ or KT5823 (a cGMP-dependent protein kinase inhibitor). However, superoxide had no effect on the basal levels of cAMP or the forskolin-induced cAMP production and moreover, the inhibition of adenylyl cyclase or cAMP-dependent protein kinase did not affect the superoxide-enhanced IP3formation. These data, therefore, suggest that the reduced cGMP formation by superoxide probably contributes to the superoxide induced activation of 1P3 formation by lifting the inhibitory feedback of cGMP on the PLC pathway(s), whereas, the cAMP pathway may not be involved in the superoxide-induced IP3formation. In vascular SMCs from SHR, the effects of superoxide were more potent than in SMCs from WKY, including the increase in 1P3 formation, the decrease in cGMP levels, and the superoxide-induced facilitation of the cross-talk interaction between cGMP and IP3pathways. The superoxide-induced 1P3formation was inhibited by a variety of antioxidants, including nacetylcysteine, cc-lipoic acid, melatonin and SOD, in vascular SMCs from both strains. It thus appears that the hypersensitivity of 1P3and cGMP pathways to superoxide is likely to contribute to the increased vascular tone and reactivity of SMCs in hypertension. Whether the effect of melatonin is due to its antioxidant properties was also explored. A greater inhibitory effect of melatonin on the norepinephrine (NE)-induced aortic contraction was observed in SHR than in Wistar-Kyoto rats (WKY). The inhibition of the NE-induced IP formation by melatonin was also greater in aortic SMCs from SHR than that from WKY. The enhanced effects of melatonin in SHR, which were found to be similarly enhanced with SOD but not with catalase, were not mediated by melatonin receptors or oc-adrenoceptors. These results indicate that the anti-hypertensive effects of melatonin are largely due to the scavenging of superoxide, and that the levels of endogenous antioxidants may not counteract the levels of overproduced superoxide in SHR. In conclusion, this study reveals a variety of novel signalling mechanisms for superoxide. For the first time, it was demonstrated that superoxide activates the hydrolysis of phosphoinositides and increases IP3levels in vascular SMCs mainly through the stimulation of tyrosine kinase-link PLCy signal pathway. It was also found that superoxide reduces cGMP formation and suppresses the cross-inhibition of IP3by cGMP, thus facilitating 1133formation. The selective effects of superoxide on 1133and cGMP pathways as well as the existence of a cross-inhibition of IP3formation by cGMP pathway provide novel mechanisms for the signalling role of superoxide in vascular SMCs. Therefore, the altered signalling effects of superoxide on the IP3pathway and the cGMP pathway, which were demonstrated in vascular SMCs from SHR for the first time in this study, could thus be responsible for the alterations in cellular signal transduction mechanisms in SHR and might contribute to the development and/or maintenance of hypertension. These observations could provide new avenues for the development of new strategies for the prevention or treatment of hypertension.