In physiological conditions, endothelial cells produce superoxide anions (O2•-) and nitric oxide (NO). O2•- are precursors, among others, of hydrogen peroxide, involved in signaling processes leading to the regulation of gene expression. At higher concentrations, hydrogen peroxide promotes the formation of the cytotoxic hydroxyl radicals. NO, in turn, exerts well-recognized beneficial effects on the cardiovascular system by inducing in particular the relaxation of smooth muscle cells. But when the NO concentration becomes too high, it competes with superoxide dismutases and reacts with O2•- to form peroxynitrite. The latter is a highly reactive molecule, attacking fatty acids, nucleic acids and proteins (with the formation of, for instance, 3-nitrotyrosine) and involved in endothelial dysfunction, a phenomenon at the basis of cardiovascular diseases. Reactive oxygen or nitrogen species can therefore, depending on the conditions, act as second messengers at low concentrations or as toxic molecules inducing oxidative stress at higher concentrations. Under normal conditions, the production of peroxynitrite is low and the damage that it could cause is limited by various defense systems, such as the Nrf2 (nuclear transcription factor erythroid 2p45 - related factor) and the UPR pathways (Unfolded Protein Response). Nrf2 is a transcription factor that responds to oxidative stress and regulates the expression of genes encoding detoxifying or antioxidant proteins such as heme oxygenase-1 (HO-1) or NAD(P)H quinone oxidoreductase 1 (NQO1). The UPR is an intracellular signaling pathway activated by the accumulation of misfolded proteins in the endoplasmic reticulum leading to the activation of various proteins such as the PERK kinase and the ATF6 transcription factor. The UPR aims to increase the synthesis of chaperonnes including BiP and GRP94, and thus increase the cellular capacity to fold or eliminate misfolded proteins. However, if the stress is too intense or too long, the UPR can induce cell death by apoptosis in particular via the activation of the transcription factor CHOP (CAAT/ enhancer binding protein (C/EBP) homologous protein). The objective of this thesis was to better understand the effects of peroxynitrite on endothelial cells, by discriminating, on one hand, its role as second messenger triggering defense responses to stress and, on the other hand, its toxicity leading to apoptosis. To achieve this goal, we chose an experimental system for peroxynitrite formation, SIN-1 (3-morpholinosydnonimine). We also tried to develop “more physiological” systems of peroxynitrite generation in vitro. Among the systems described to generate peroxynitrite, we chose homocysteine, hypoxia/reoxygenation, oxidized LDL and angiotensin-II. In this study, we developped an experimental model of endothelial cell exposure to peroxynitrite by incubating human endothelial cells in culture (the EAhy926 cell line and the primary culture, HUVEC) in the presence of SIN-1. Firstly, we tested the cytotoxicity of SIN-1 and we characterized the peroxynitrite formation induced by SIN-1 in these cellular models by following the oxidation of the hydroxyphenyl fluorescein probe and the formation of 3-nitrotyrosine by Western blot with anti-3-NT antibodies. Secondly, we demonstrated the activation of the Nrf2 and UPR pathways by SIN-1. We also showed that SIN-1 was able to exert a protective effect in serum-starved EAhy926 cells. By investigating some of the molecular mechanisms involved in the cytoprotective effect of SIN-1, we found that, via Nrf2 and HO-1 proteins, SIN-1 induced a decrease in DNA fragmentation and increased LC3-II formation in serum-starved endothelial cells. We also evaluated the effects of more physiological conditions (homocysteine, hypoxia/reoxygenation, oxidized LDL and angiotensin-II) on the peroxynitrite formation. It seems that the latter is generated in the presence of oxidized LDL and angiotensin-II combined with a NO donor. However, more investigations are necessary to further optimize some of these “more physiological” conditions on the peroxynitrite formation and to confirm the cytoprotective effect of peroxynitrite observed in serum-starved endothelial cells exposed to SIN-1. Altogether, our results indicate that in a narrow range of concentration, peroxynitrite formed after SIN-1 stimulation, activates the Nrf2 and UPR pathways, leading to a cytoprotective effect in serum-starved endothelial cells. This protective effect is achieved by stimulating the endothelial cells either after or before a period of serum starvation. The latter is in agreement with so called conditioning experiments in vivo. This work provides some new insights in the balance between survival signals or cell death triggered by peroxynitrite and highlights under which conditions this balance leads to endothelial cell death, in the context of endothelial dysfunction, involved in the early stages of atherosclerosis. It also suggests that a drastic anti-oxidant therapy targetting cardiovascular diseases could have drawbacks, as it abolishes the positive effects of reactive oxygen and nitrogen species.