What is the topic of this review? The review concerns the role of reactive oxygen species as physiological second messengers in potentiating G-protein-coupled receptor-mediated vasoconstriction and its potential dysregulation by oxidant stress in pulmonary hypertension. What advances does it highlight? The review highlights the concept that physiological signalling by reactive oxygen species must normally be highly compartmentalized to prevent self-regenerating oxidant stress and promiscuous and uncontrolled signalling, which contribute to the aetiology. Pulmonary hypertension is associated with oxidant stress and increased generation of reactive oxygen species (ROS) by NADPH oxidases (NOX), mitochondria and other sources. There is considerable evidence that these contribute to the aetiology via promotion of pulmonary vascular remodelling, endothelial dysfunction and enhanced vasoreactivity. However, it is now recognized that ROS act as important signalling mediators and second messengers in normal physiological conditions. Many ion channels and protein kinases crucial to pulmonary vascular function are directly or indirectly affected by redox/ROS, including K , Ca and non-selective cation channels and Rho kinase. However, the inherent difficulties in quantifying ROS, particularly in subcellular compartments, make it uncertain whether these reported effects are of relevance in physiological rather than pathological conditions. In an attempt to address such issues, we have focused on the role of physiologically generated ROS in the regulation of G-protein-coupled receptor (GPCR)-activated vasoconstrictor pathways. We have recently reported a novel mechanism whereby low concentrations of GPCR-linked vasoconstrictors greatly potentiate Ca entry via a NOX1- and ROS-mediated pathway parallel to the classical vasoconstrictor pathways of Ca mobilization and activation of Rho kinase. Our findings imply that ROS signalling is highly compartmentalized in physiological conditions, but that this may be compromised by pathological increases in oxidant production, for example in pulmonary hypertension, leading to promiscuous actions that contribute to the aetiology. This model is consistent with the proposal that targeted antioxidants could prove to be an effective therapy for pulmonary hypertension.