Is hydrogen sulfide a circulating "gasotransmitter" in vertebrate blood?

Hydrogen sulfide (H(2)S) is gaining acceptance as a signaling molecule and has been shown to elicit a variety of biological effects at concentrations between 10 and 1000 micromol/l. Dissolved H(2)S is a weak acid in equilibrium with HS(-) and S(2-) and under physiological conditions these species, collectively referred to as sulfide, exist in the approximate ratio of 20% H(2)S, 80% HS(-) and 0% S(2-). Numerous analyses over the past 8 years have reported plasma or blood sulfide concentrations also in this range, typically between 30 and 300 micromol/l, thus supporting the biological studies. However, there is some question whether or not these concentrations are physiological. First, many of these values have been obtained from indirect methods using relatively harsh chemical conditions. Second, most studies conducted prior to 2000 failed to find blood sulfide in micromolar concentrations while others showed that radiolabeled (35)S-sulfide is rapidly removed from blood and that mammals have a relatively high capacity to metabolize exogenously administered sulfide. Very recent studies using H(2)S gas-sensing electrodes to directly measure sulfide in plasma or blood, or HPLC analysis of head-space gas, have also indicated that sulfide does not circulate at micromolar levels and is rapidly consumed by blood or tissues. Third, micromolar concentrations of sulfide in blood or exhaled air should be, but are not, malodorous. Fourth, estimates of dietary sulfur necessary to sustain micromolar levels of plasma sulfide greatly exceed the daily intake. Collectively, these studies imply that many of the biological effects of sulfide are only achieved at supra-physiological concentrations and they question whether circulating sulfide is a physiologically relevant signaling molecule. This review examines the blood/plasma sulfide measurements that have been reported over the past 30 years from the perspective of the analytical methods used and the potential sources of error.