Inhaled nitric oxide (NO) is a pulmonary vasodilator, but also acts systemically, causing negative cardiac inotropic effects and a fall in systemic vascular resistance. Circulating metabolites of NO are presumed to be responsible. We questioned the role of nitrite anions and the manner in which they might contribute to these effects. Nitrite and nitrate anions coexist in blood, while circulating levels of dissolved NO are very low. Nitrate anions are not biologically active, but nitrite anions may have a biological role through the release of NO. In vitro, at 37 degrees C and in aerated Krebs bicarbonate solution, the steady-state concentration of dissolved NO was proportional to the concentration of NO in the gas. Nanomolar concentrations of dissolved NO coexisted with micromolar concentrations of nitrite anions. The idea of an equilibrium between the two in solution was also supported by the observed release of NO from nitrite anions in the absence of gas. With rings of precontracted pig pulmonary arteries (prostaglandin F(2alpha); 10 micromol/l), the steady-state concentration of dissolved NO causing 50% relaxation (EC(50)) was 0.84+/-0.25 nmol/l, corresponding to a gaseous concentration of 2.2 p.p.m. The EC(50) of nitrite was 4.5+/-0.7 micromol/l, a concentration normally found in plasma. The estimated concentration of dissolved NO derived from this nitrite was 4.5 pmol/l, some 100 times lower than would be needed to cause relaxation. The rate of exhalation of NO was increased and pulmonary vascular resistance was reduced by the addition of nitrite solution to the perfusate of isolated perfused and ventilated pig lungs, but only when millimolar concentrations were achieved. Thus circulating nitrite anions are a direct vasodilator, only being a carrier of effective amounts of "free" NO at higher than physiological concentrations.