© 2005 American Thoracic Society DOI: 10.1165/rcmb.F294
Hemoglobin as a Nitrite Reductase Regulating Red Cell–Dependent Hypoxic VasodilationNational Institutes of Health Bethesda, Maryland Hypoxic vasodilation is a highly conserved physiologic vasodilatory response to low tissue oxygen tension that ensures delivery of blood to match metabolic demand. Although this response has been appreciated for more than 120 yr and characterized in the coronary circulation for 80 yr, the mechanism of oxygen and pH sensing and the effectors of vasodilation remain uncertain (recently reviewed [1, 2]). The classical paradigm suggests that the oxygen sensor detects the partial pressure of oxygen in tissue and responds with feedback release of local mediators such as adenosine, KATP channel activation, eNOS-derived NO, and/or prostacyclin, as well as feed-forward sympathetic activation of ß-adrenergic receptors. However, experimental blockade of these pathways does not completely block hypoxic vasodilation, supporting the existence of either overlapping pathways or unknown additional mediators (1, 3). Recent studies have proposed a new paradigm for hypoxic vasodilation in which hemoglobin's allosteric conformational state serves as the oxygen sensor with the allosteric quaternary structural transition from R (relaxed or oxygenated tetrameric conformation) to T (tense or deoxygenated tetrameric conformation), triggering the release of either erythrocyte ATP (4–6), S-nitrosothiol from S-nitrosated hemoglobin (SNO-Hb) (7), or a reduction of the anion nitrite to NO by deoxyhemoglobin (8, 9). In the review in the April issue of AJRCMB, Robinson and Lancaster comprehensively review the data and controversies surrounding the latter two mechanisms, the SNO-hemoglobin hypothesis and the nitrite reductase hypothesis, and highlights the unique challenges posed by each model (10). It is certain that more experiments are required and years of work by many laboratories are necessary to establish the validity of the two models and the relative contributions, if any, of either process to physiologic blood flow control and vascular NO-dependent homeostasis (controversies recently reviewed [11]). However, increasing data suggest that the simple ubiquitous anion nitrite represents a potentially large reservoir of NO that serves as a critical hypoxic buffer, potentially regulating mitochondrial respiration, ischemia-reperfusion, and hypoxic vasodilation (8, 9, 12–18). Central to these observations is the recent appreciation that hemoglobin, certainly the most studied protein in human history, and potentially other heme proteins such as myoglobin, cytoglobin, and neuroglobin, may have important physiologic roles as "enzymatic" nitrite reductases (8, 9, 15, 19). From a historical standpoint, data from as early as 1930 suggested that large doses of nitrite vasodilate the mammalian circulation, an observation that limited the dosing of nitrite as an antidote for cyanide poisoning. Robert Furchgott used nitrite as a vasodilator to develop the aortic ring bioassay in 1952, and both Murad and Ignarro explored the effects of nitrite on guanylate cyclase activation (20–22). However, the high concentrations of nitrite required to vasodilate aortic ring bioassay systems at room temperature and normal oxygen tension led to a premature dismissal of nitrite as a physiologic vasodilator (23–25). In our work exploring possible endocrine mediators of NO-dependent vasodilation, we evaluated arterial-venous gradients of NO-modified proteins and oxidation products in the circulation and found significant gradients in nitrite, with increased consumption with exercise stress (an observation we and others could never reproduce for SNO-Hb) (17). Based on mechanisms suggesting that nitrite might be bioconverted to NO by xanthine oxidoreductase (18, 26) or disproportionation (12), we hypothesized that nitrite might vasodilate the human circulation under exercise stress. We were thus surprised to find that nitrite vasodilated the human circulation even under normal physiology at concentrations as low as 900 nM (8). This vasodilation was associated with the rapid formation of iron-nitrosyl-hemoglobin across the forearm circulation that was inversely proportional to the deoxyhemoglobin concentration in blood perfusing this vascular bed. This observation suggested a novel mechanism of nitrite bioactivation based on a simple reaction of nitrite with deoxyhemoglobin as described by Doyle and colleagues (19).
We were struck by the potential physiologic implications of this simple equation. It requires deoxyhemoglobin and a proton, providing oxygen and pH sensor chemistry, and generates NO, a potent vasodilator, and methemoglobin, which will not autocapture and inactivate the NO formed. In additional experiments we found that nitrite, red cells (or hemoglobin), and hypoxia were required for in vitro hypoxic vasodilation. Indeed, in the presence of hypoxia and erythrocytes (conditions never tested in historical aortic ring bioassay studies), nitrite now vasodilated aortic rings at physiologic concentrations (8). In ongoing work from four laboratories (the Gladwin, Patel, Kim-Shapiro, and Hogg groups), we are finding that this vasodilation occurs as hemoglobin unloads oxygen to 50% saturation, and that this vasodilation is mediated by a maximal nitrite reductase activity of hemoglobin allosterically linked to its intrinsic P50. Such a set point is remarkable as classical hypoxic vasodilation occurs at this hemoglobin oxygen saturation (between 40 and 60%) (27). I would like to briefly address the current major challenges facing the "nitrite reductase hypothesis" in the hopes that these questions are taken on by multiple laboratories and that accumulated data will advance the field beyond the current state of controversy. Does Nitrite-Dependent Vasodilation Play a Role in Normal Physiologic Control of Blood Flow and Vasodilation? There exist artery-to-vein gradients of nitrite in the normal human circulation with increased consumption of nitrite with regional metabolic stress (17). Such an observation is a necessary but insufficient requirement for a feedback mediator of hypoxic vasodilation. Using an in vitro aortic ring bioassay system designed by the Patel lab to simultaneously measure vessel force tension and oxygen tension, we find that vasodilation is measureably potentiated by as low as 200 nM nitrite under hypoxic conditions (8). In humans, we observe significant vasodilation at 900 nM (increases in blood flow of 22%) (8) and in other animal models, such as the anesthesized canine, we have observed profound vasodilation (decreases in mean arterial blood pressure of 5–10 mm Hg) at systemic levels of only 5–8 µM (unpublished observations). Interestingly, cannine erythrocytes do not release ATP, possibly suggesting a greater dependence and sensitivity to nitrite-mediated hypoxic vasodilation (R. Sprague, personal communication). Additional studies have been published in the last year confirming the vasodilating effects of nitrite (15, 28–30). Although these results suggest that nitrite is surprisingly potent and has the potential to participate in vascular NO homeostasis and hypoxic vasodilation at the physiologic levels of 300 nM measured in red blood cells, we need to devise a way to inhibit or inactivate circulating nitrite to fully test this in vivo. Is the Degree of Deoxygenation in the Precapillary Microcirculation Sufficient for Nitrite-Deoxyhemoglobin–Mediated Vasodilation? Nitrite appears to best fit the current requirement for a physiologic mediator of hypoxic vasodilation, as it maximally reacts with hemoglobin at 60–40% hemoglobin saturation, an oxygen tension (40–20 mm Hg) significantly higher than that required for SNO-hemoglobin deoxygenation (secondary to the high affinity of cysteine 93 liganded hemoglobin [31, 32]). In the normal skeletal muscle circulation, oxygen tension decreases from the A1 caliber arterioles (100 µm diameter) to the A4 caliber arterioles (20 µm diameter) to values as low as 20 mm Hg before the capillary circulation (recently reviewed [2]). These data suggest that most oxygen delivery occurs within the arterioles, allowing for spacially linked oxygen delivery and vasomotor control. Additional mechanisms suggest that acetylcholine, NO, or ATP delivery to capillary circulation produces retrograde intracellular propagation of vasodilating signal to the precapillary resistance vessels (33–35). Thus a maximal nitrite reductase activity at the hemoglobin P50 appears ideal for oxygen sensing and hypoxic vasodilation, as this allosteric point is thermally, chemically, and electronically responsive to physiologically relevant tissue metabolic stress. How Does NO Escape the Erythrocyte? Is There an Intermediate Gas Species Other than NO? As described in the review by Drs. Robinson and Lancaster (10), this question poses the major challenge for the nitrite reductase hypothesis and represents a major area of investigation by our lab and collaborators. Based on the kinetics of NO reaction with intracellular deoxyhemoglobin, NO should not be able to escape the erythrocyte following formation from nitrite. However, we do detect the ultimate formation of NO in our experiments:
Although NO clearly mediates the nitrite effect, how NO can escape remains uncertain, and a number of possible explanations are being tested. For example, membrane-associated nitrite reductase metabolons with deoxyhemoglobin and methemoglobin, anion exchange protein, carbonic anhydrase, aquaporin, and Rh channels will bring together nitrite, proton, deoxyheme, and highly hydrophobic channels that could serve to concentrate the lipophilic NO at the membrane complex (36). The potency of NO (EC50 of only 1–5 nM) requires very little NO escape to regulate vasodilation (flow is proportional to the radius to the fourth power!). Other potential intermediates that may form in the nitrite deoxyhemoglobin reaction including H2NO2 (hydrated NO), HNO2–, NO2–2 (described in Ref. 19), NO, and N2O3 (N2O3 formation by reductive nitrosylation catalyzed by nitrite has recently been characterized by Fernandez and colleagues [37]). Many of these intermediates would be stabilized in hydrophobic channels (H2NO2, N2O3, and NO), whereas others in anion channels such as the anion exchange protein (HNO2– and NO2–2). In conclusion, more research is required to address remaining questions presented by the nitrite reductase hypothesis, such as the role of nitrite in normal physiology and potential novel intermediate chemical species formed in the reaction; it is our hope that other laboratories take on this challenge to help resolve these questions of fundamental biological importance. Footnotes Conflict of Interest Statement: M.T.G. has no declared conflicts of interest. References
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
