Background: Pulmonary hypertension (PH) is a rare condition characterized by endothelial dysfunction and vascular remodelling, leading to increased pulmonary vascular resistance (PVR) and right ventricular heart failure. Endothelial dysfunction is associated with an imbalance between vasoconstrictor compounds, such as endothelin and thromboxane A2, and vasodilator compounds, such as prostacyclin and nitric oxide (NO). Asymmetric dimethylarginine (ADMA), a methyl derivate of L-arginine, inhibits synthesis of NO. Vardenafil, a phosphodiesterase type 5 inhibitor (PDE5-inhibitors), causes vasodilation through the NO/cGMP pathway. Aim: This thesis investigates the pharmacological effects and diagnostic utility of vardenafil in PH patients. In addition, to evaluate the change of L-arginine and dimethylarginines before and during PAHspecific therapy in PAH patients compared to patients with left ventricular heart failure (LVHF) and healthy subjects. Methods: The pharmacokinetics and hemodynamic effects of vardenafil were examined during right heart catheterization (RHC) in 16 PH patients and plasma concentrations were measured for up to nine hours after oral administration. In 20 PH patients, acute vasoreactivity test with vardenafil was performed during RHC. Hemodynamic responses were recorded, responders were defined and followed for up to seven years. Additionally, plasma ADMA, symmetric dimethylarginine (SDMA), L-arginine, L-citrulline and L-ornithine levels before and after PAH drug treatment were monitored in 21 PAH patients and compared to values measured in 14 LVHF patients and 27 healthy subjects. Results: Vardenafil concentrations increased rapidly to maximum plasma concentration (tmax 1h) and elimination half-life was 3.4 h. Patients co-medicated with bosentan had reduced vardenafil concentration. Significant acute hemodynamic responses were observed for mean pulmonary artery pressure (mPAP) (p<0.001), pulmonary vascular resistance (PVR) (p<0.001), cardiac output (CO) (p=0.015), cardiac index (CI) (p=0.010), systemic vascular resistance (SVR) (p<0.001) and PVR/SVR (p=0.002) and were related to plasma vardenafil concentrations. PAH patients had significantly higher ADMA and SDMA levels and significantly lower L-arginine levels and L-arginine/ADMA ratio compared with healthy subjects (p<0.001). L-arginine was also lower in PAH patients compared to patients with LVHF (p<0.05). WHO functional class and six minutes walking distance (6MWD) correlated to Larginine and L-arginine/ADMA ratio in PAH at baseline (p<0.05). At follow-up, patients on mono- or combinationtherapy with endothelin receptor antagonists (ERA) had lower ADMA levels than patients without ERA (p<0.05). In contrast, patients on PDE5-inhibitors had higher ADMA levels compared to patients without PDE5-inhibitors (p<0.05). Conclusion: Vardenafil is safe in acute vasoreactivity test in PH patients. Cardiopulmonary hemodynamic response was related to plasma drug concentrations. There was a high inter-individual variability of vardenafil pharmacokinetics and co-medication with bosentan caused a pharmacokinetic drug interaction. Baseline L-arginine and dimethylarginines levels were different in PAH patients compared to LVHF patients and healthy controls. PAH-specific treatment influenced L-arginine and dimethylarginines. Our data suggest that L-arginine might be useful for differentiating PAH from LVHF, and L-arginine/ADMA ratios were related to the severity of PAH and might be useful for follow-up evaluations of PAH patients.

La cholestase intrahépatique est une manifestation fréquente des lésions hépatiques induites par les médicaments; Cependant, les mécanismes impliqués sont peu connus. Nous avons cherché à étudier les mécanismes de la cholestase induite par les médicaments afin d’améliorer sa détection précoce en utilisant les cellules HepaRG humaines. Tout d'abord, nous avons prouvé que les canalicules biliaires (BC) subissaient des contractions spontanées, essentielles pour l'efflux d’acides biliaires et nécessitaient des séries d’alternance dans la phosphorylation/déphosphorylation de la chaîne légère de myosine (MLC2). La courte exposition à des composés cholestatiques a révélé que la modulation des BC était associée à des perturbations de la voie de signalisation ROCK/MLCK. Afin de confirmer notre étude, 12 médicaments cholestatiques et six non cholestatiques ont été analysés et nous avons démontré que tous les médicaments cholestatiques classés sur la base de résultats cliniques provoquaient des perturbations dans la dynamique des BC (dilatation ou constriction), et altéraient la voie de signalisation ROCK/MLCK, tandis que les composés non cholestatiques n'avaient pas d'effet. Nous avons également prouvé que ces changements étaient plus spécifiques que la mesure de l'inhibition de l’efflux comme marqueurs prédictifs non cliniques de la cholestase induite par les médicaments. Afin de confirmer et d’étendre ces conclusions, nous avons analysé les mécanismes impliqués dans les effets cytotoxiques et cholestatiques induits par 4 médicaments de la famille des antagonistes des récepteurs de l'endothéline: deux ayant un lien avec des cas cliniques d'hépatotoxicité (sitaxentan) et/ou cholestase (bosentan), et deux n’ayant pas été impliqués dans l’élévation de transaminases hépatiques ou de bilirubine (ambrisentan et macitentan). Les résultats montrent que le macitentan récemment commercialisé et ayant une structure chimique similaire à celle du bosentan, était capable de causer, comme ce dernier, des altérations in vitro des BC. En revanche, l'ambrisentan n’était pas hépatotoxique et le sitaxentan qui a été retiré du marché pour des cas d’hépatotoxicité, n’affectait pas la dynamique canaliculaire. / Intrahepatic cholestasis represents a frequent manifestation of drug-induced liver injuries; however, the mechanisms involved in such injuries are poorly understood. We aimed to investigate mechanisms underlying drug-induced cholestasis and improve its early detection using human HepaRG cells. First, we proved that bile canaliculi (BC) underwent spontaneous contractions, which are essential for bile acid efflux and required alternations in myosin light chain (MLC2) phosphorylation/dephosphorylation. A short exposure to cholestatic compounds revealed that BC dynamics was altered and associated with impairment of the ROCK/MLCK pathway. Then, in order to confirm our study, 12 cholestatic drugs and six noncholestatic drugs were analyzed and we demonstrated that all cholestatic drugs classified on the basis of reported clinical findings caused disturbances of both BC dynamics (dilatation or constriction), and alteration of the ROCK/MLCK signaling pathway, whereas noncholestatic compounds had no effect. We also proved that these changes were more specific than efflux inhibition measurements alone as predictive nonclinical markers of drug-induced cholestasis. To confirm and extend these conclusions, we analyzed the mechanisms involved in cytotoxic and cholestatic effects induced by the 4 main drugs from the endothelin receptor antagonists family: two related to clinical cases of hepatotoxicity (sitaxentan) and/or cholestasis (bosentan), and two that have not been reported to cause elevation of liver transaminases or bilirubin (ambrisentan and macitentan). The results showed that like bosentan, the structurally similar recently marketed drug, macitentan, could cause in vitro major BC alterations. By contrast, ambrisentan appeared as a safe drug and sitaxentan that has been withdrawn from the market for hepatotoxic cases, did not impair BC dynamics.