In this special issue of the Journal of Molecular Medicine, we present five review articles concerning small molecules that play big roles in physiology and medicine: the gases oxygen (O2), nitric oxide (NO), carbon monoxide (CO), and hydrogen sulfide (H2S). Of these, the essential requirement for O2 is well known to physicians, scientists, and laymen alike. However, it is only within the last two decades that we have begun to understand the molecular mechanisms by which every cell in our body senses the local O2 concentration and responds to reduced O2 availability (i.e., hypoxia) with changes in gene expression that are mediated by the hypoxia-inducible factors HIF-1 and HIF-2 [1]. Recent evidence suggest that NO, CO, and H2S function as signaling molecules that also play critical roles in regulating O2 homeostasis.

Philip Marsden and colleagues (University of Toronto) describe the fascinating crosstalk between O2 sensing and NO signaling that occurs to regulate red blood cell levels and blood vessel tone, which together play critical roles in O2 delivery [2]. In particular, they discuss exciting data from their lab demonstrating that the physiological responses to anemic hypoxia (reduced O2 carrying capacity) and hypoxic hypoxia (reduced O2) are mechanistically distinct. This difference is dramatically illustrated by their finding that subjecting mice that lack neuronal NO synthase (also known as nNOS or NOS1) to anemic hypoxia resulted in an increased mortality rate, compared to wild-type mice, whereas these NOS1-deficient mice were protected against mortality during hypoxic hypoxia [3]. Their delineation of the molecular and cellular basis for these effects is truly elegant physiology. The crosstalk between O2 sensing and NO signaling is extensive and includes increased expression of inducible NOS (also known as iNOS or NOS2) under hypoxic conditions that is mediated by HIF-1 [4, 5] and the regulation of HIF-1 activity by S-nitrosylation of a cysteine residue in the HIF-1α subunit [6].

Puneet Anand and Jonathan Stamler (Case Western Reserve University) provide a global view of protein S-nitrosylation and its effects on protein function [7]. They describe several different molecular mechanisms by which proteins are nitrosylated and counteracting mechanisms by which they are denitrosylated, which is analogous to the phosphorylation and dephosphorylation of proteins by kinases and phosphatases. The balance between nitrosylation and denitrosylation is dependent on the redox state of the cell, thus establishing inherent crosstalk between NO and reactive oxygen species. The authors discuss several examples of diseases associated with dysregulated S-nitrosylation, which may therefore represent a novel therapeutic target.

Makoto Suetmatsu and colleagues (Keio University) discuss the biological role of CO, which is generated by heme oxygenases (HO1 and HO2) using heme and O2 as substrates [8]. They discovered that CO binds to the heme moiety of cystathionine β-synthase (CBS), an enzyme that generates H2S, and that CO inhibits CBS catalytic activity [9]. More recently, an elegant series of experiments have revealed that in the brain, CO is generated by HO2, binds to CBS, and inhibits its activity; however, under hypoxic conditions, CO production falls, CBS is no longer inhibited, resulting in the production of H2S, which functions as a vasodilator [10]. Thus, changes in O2 availability result in adaptive changes in cerebral vascular tone (and tissue perfusion) at least in part through changes in HO2- and CBS-dependent production of CO and H2S, respectively.

Scott Vandiver and Solomon Snyder (Johns Hopkins University) focus on the mechanisms and consequences of H2S production [11]. In addition to CBS, H2S is also generated by cystathionine γ-lyase (CSE) and 3-mercaptopyruvate sulfotransferase. H2S mediates biological effects by sulfhydration of thiol groups of cysteine residues within proteins, including the glycolytic enzyme glyceraldehyde-3-phosphate dehydrogenase, leading to increased enzyme activity [12]. Sulfhydration of ATP-dependent potassium channels mediates the vasodilation described above [13]. Just as Anand and Stamler review the evidence that protein nitrosylation mediates the effects of NO, Vandiver and Snyder show that sulfhydration is emerging as a novel and widespread posttranslational modification that mediates the effects of H2S.

In the final paper, we describe how NO, CO, and H2S are all involved in O2 sensing and signal transduction by the carotid body, a small sensory organ located at the bifurcation of the common carotid artery [14]. Glomus cells within the carotid body sense the arterial O2 concentration and transduce neural signals to the brain centers that control respiration and blood pressure in order to increase tissue oxygenation. Remarkably, carotid bodies from mice that are heterozygous for a knockout allele at the locus encoding HIF-1α do not respond to hypoxia [15], whereas carotid bodies from mice that are heterozygous for a knockout allele at the locus encoding HIF-2α show exaggerated responses to hypoxia [16], providing evidence that the homeostatic responses mediated by the carotid body are dependent upon a balance between HIF-1 and HIF-2. Glomus cells express CSE and carotid bodies from mice lacking CSE also do not respond to hypoxia [17], whereas NOS1-deficient mice exhibit exaggerated ventilatory responses to hypoxia [18]. Treatment with a CO donor blocks hypoxia-induced H2S production, suggesting that CO produced by HO2 inhibits CSE activity in the carotid body [17], although the mechanism of action is unknown because, unlike CBS, CSE does not contain a heme group to which CO can bind. The targets of sulfhydration by H2S in the carotid body also remain to be determined.

Taken together, these review articles demonstrate that O2, NO, CO, and H2S play critical roles in a wide range of fundamental homeostatic mechanisms, which become dysregulated in various disease processes. It is likely that future studies will uncover many more clinical contexts in which the action of these small molecules is relevant and may be targeted therapeutically.