Current models of circulatory control postulate that the red blood cell (RBC) acts as an O2 sensor and regulator of local muscle blood flow at rest and during exercise in humans (González-Alonso et al. 2001) via three signalling mechanisms: 1) ATP release in proportion to the degree of oxyhaemoglobin saturation (FO2Hb) (Ellsworth et al. 1995); 2) S-nitrosohaemoglobin (SNO-Hb)-dependent vasodilatation (Stamler et al. 1996), and 3) reduction of plasma nitrite to vasoactive NO by deoxyhaemoglobin (Cosby et al. 2003). Whether these local signals also mediate the cardiac output (Q) response is presently unknown. Thus, the aim of this study was to examine whether alterations in leg blood flow (LBF) and Q with graded hypoxia and hyperoxia at rest and during exercise are intimately related to changes in FO2Hb and alterations in either plasma ATP, SNO-Hb and/or nitrite. To this aim, we measured leg and systemic haemodynamics, O2 transport and arterial and femoral venous haematological variables together with SNO-Hb, ATP and nitrite at rest and during 6-min constant-load knee-extensor exercises (25±1 W or ~30% of peak power; mean±SEM) in 7 active male subjects (27±2 yr) in normoxia and under 10%, 13%, 16%, 48% and 100% FiO2. Data were analysed by a one-way ANOVA with repeated measures and Pearson product moment correlations. By design, arterial FO2Hb at rest and during exercise was gradually increased from 75 to 99 % and 70 to 99 %, respectively. LBF and Q were linearly related to changes in arterial FO2Hb at rest (r=-0.45 and r=-0.77; P<0.05) and during exercise (r=-0.39 and r=-0.69; P<0.05), such as leg and systemic O2 delivery was kept constant across conditions. Despite the large changes in FO2Hb at rest and during exercise, arterial and venous plasma ATP, SNO-Hb and nitrite remained unchanged. However, when all conditions are taken into account, lower FO2Hb correlated positively with plasma nitrite (r=0.27; P<0.05) and negatively with SNO-Hb (r=-0.29; P<0.05). At rest and during exercise, LBF and Q correlated with arterial plasma nitrite (r=-0.34 to -0.66) whereas during exercise LBF also correlated with venous plasma nitrite (r=-0.60; P<0.05). In conclusion, limb muscle and systemic blood flow are closely tied to blood oxygenation over a wide range of O2 availability. Few changes appeared in plasma ATP, SNO-Hb and nitrite, possibly because any increase in their concentrations might be followed by rapid degradation, utilisation or scavenging. Nevertheless, plasma nitrite were coupled to FO2Hb and LBF suggesting that the consumption of nitrite may be essential to local muscle flow regulation in which SNO could be a marker of nitrite reduction by the erythrocyte.