Hocus pocus hypoxia – NO and augmented vasodilatation in the systemic vasculature during hypoxic exercise

Evidence clearly points to an enhanced vasodilatation in systemic vessels during hypoxic exercise that serves to defend oxygen delivery to active musculature in the face of a reduced inspired fraction of O₂ (Wilkins et al. 2008). Exercise-induced hyperaemia is a complex process with redundant mechanisms that can be called upon when required. During hypoxic exercise there is an additional dilator response in human skeletal muscle attributable to the reduction in arterial oxygen content rather than arterial O₂ tension per se. Importantly, prevailing vascular tone results from both neural and metabolic factors acting on the vascular smooth muscle and endothelium. At the vascular endothelium adrenergic and non-adrenergic vasoactive pathways play a regulatory role in normal vascular function. The overall effect on haemodynamics and arterial pressure will be determined by how these various pathways and mechanisms integrate at the level of the vascular smooth muscle cell. Over recent years various research groups, including ours (Bailey et al. 2009), have combined a number of invasive experimental protocols across isolated vascular beds in conjunction with pharmacological blockade of vasoactive metabolite receptors or the actual metabolite itself to elucidate any potential contribution to exercise and/or hypoxic vasodilatation.

One of the most intensely studied candidates for both exercise hyperaemia and hypoxic vasodilatation is nitric oxide (NO^•). Increases in blood flow, cyclic wall stress due to pulsatile blood flow and catecholamines produce an up-regulation and release of NO^• from the vascular endothelium (Busse & Fleming, 2006) via the enzyme endothelial nitric oxide synthase (eNOS). Hypoxia has been associated with additional sources of NO^• release from deoxyhaemoglobin, β-adrenergic and adensosine receptor stimulation (Stamler et al. 1997; Bryan & Marshall, 1999; Wilkins et al. 2008). The NO^• released toward the vascular lumen is a powerful vasodilator responsible for mediating basal vascular tone (Stamler et al. 1997). However, not all vascular beds respond in a similar manner with the pulmonary vasculature demonstrating a strong hypoxia-induced vasoconstriction whereas the cerebral vasculature responds in a similar fashion to the systemic vessels with a vasodilatation (Bailey et al. 2009). Metabolism of NO^• within the vasculature to the more biochemically stable moiety nitrite serves as a means to determine circulating bioavailability of NO^•. It appears that whilst this metabolic pathway of NO^• was initially considered unidirectional, exogenous nitrite can induce sustained vasodilatation especially when the local vascular environment is hypoxic or ischaemic (Maher et al. 2008). It is within this environment that deoxygenated haemoglobin appears to convert nitrite to NO^• (Stamler et al. 1997).

Our laboratory, in collaboration with others, recently reported a reduced pulmonary vasoconstriction with systemic infusion of sodium nitrite (Ingram et al. 2010) while others have also reported augmented systemic arterial hypoxic vasodilatation with the same agent (Maher et al. 2008). Thus, with this background it is evident that during hypoxic exercise there is a compensatory vasodilatation that is sustained during increased exercise intensity and a clear contender for mediating the response is NO^• either from enhanced endothelial release and/or circulating deoxyhaemoglobin.

Casey et al. (2010), in a recent article in the The Journal of Physiology, sought to address this issue by examining the effects of hypoxic forearm exercise whilst simultaneously infusing the NOS inhibitor N^G-monomethyl-l-arginine (l-NMMA) to investigate the influence of endothelial-derived NO^•. In a parallel branch of the investigation, the authors also attempted to glean further information regarding hypoxic-induced NO^• release via adenosine receptor stimulation by exogenous administration of combined l-NMMA and aminophylline, an adenosine receptor antagonist (Casey et al. 2010). Efficacy of eNOS blockade was established via intra-arterial acetylcholine infusion. Casey and colleagues (2010) utilised the isolated forearm exercise model with 22 healthy young adults. Subjects performed rhythmic forearm exercise in the non-dominant arm at 10% and 20% of individual maximal voluntary contraction. Twelve subjects completed protocol 1 (saline or l-NMMA infusion) and ten subjects completed protocol 2 (saline or l-NMMA–aminophylline infusion). Due to the long half-life of l-NMMA, study drugs were administered in the same order. Exercise was performed in normoxia and normocapnic hypoxia. Hypoxic inspiration rendered systemic arterial O₂ saturations at ∼80%. Arterial pressure responses were monitored with an indwelling pressure transducer in the brachial artery whilst forearm blood flow was determined in the brachial artery via ultrasound. Forearm vascular conductance was calculated by the quotient of forearm blood flow and arterial pressure (Casey et al. 2010). The paper highlights three key findings of importance regarding the role of NO^• in hypoxic vasodilatation.

NO mediates the augmented systemic vasodilatation during incremental hypoxic exercise

Casey et al. (2010) demonstrate for the first time that systemic infusion of a NOS inhibitor substantially attenuates (approximately 14% reduction across workloads) the augmented hyperaemia during hypoxic exercise of increasing intensity. Whilst β-adrenergic receptor stimulation has been implicated previously as mediating hypoxic vasodilatation, this component decreases with increased exercise intensity (Wilkins et al. 2008). Thus, NO^• appears responsible for orchestrating increases in blood flow during hypoxic exercise that is robust across increased exercise intensities. Evidence suggests that NO^•-mediated vasodilatation during hypoxia at rest may be/resultant from adenosine receptor stimulation (Bryan & Marshall, 1999). Casey et al. (2010) proceeded to evaluate for the first time if a double blockade of NOS and adenosine receptors would further attenuate the hypoxic vasodilatation during exercise.

Failure of adenosine receptor-stimulated NO^• release after NOS inhibition during incremental hypoxic exercise

The authors observed a lack of any further reduction in augmented hypoxic vasodilatation after antagonism of both eNOS and adenosine receptors. Therefore it appears that adenosine receptor activation is not a major source of NO^• production during hypoxic exercise. This finding is consistent with recent literature removing adenosine from the role of primary modulator of hypoxic vasodilatation at least during exercise conditions. With the lack of adenosine-activated NO^• release other candidates now come to the forefront. Of these candidates, whilst a strong possibility exists for a prostaglandin and NO^• interaction regulating skeletal muscle blood flow at rest and during exercise, it still remains to be determined if this relationship exists in a hypoxic milieu. One of the most likely mechanisms is ATP release through oxygen-sensitive mechanisms in erythrocytes or endothelial cells during hypoxia that mediates its effects through an NO^• pathway. With it established that NO^• is at the centre of hypoxic-mediated vasodilatation, an intriguing area of investigation with direct clinical application is the exact site of NO^• release via haemoglobin or endothelial mechanisms.

Hypoxic NO-mediated vasodilatation is endothelial by origin

A hypoxic endothelial lumen elicits direct release of NO^• via eNOS or from desaturated erythrocytes in the form of S-nitrosohaemoglobin (Stamler et al. 1997). With the choice of intraluminal NOS inhibition and the noted reductions in vasodilatation, the study by Casey et al. (2010) argues for an endothelial regulation of NO^• during hypoxic exercise rather than an erythrocytic source. By definition, this study also provides a strong case against nitrite either directly or indirectly, via reduction to NO^•, being a key vasodilator in hypoxic skeletal muscle. Under this scenario NOS inhibition would not have attenuated hypoxic vasodilatation. Whilst NOS inhibition was unselective of NOS isoform it is unlikely that neuronal NOS (nNOS) or inducible NOS (iNOS) played a role in the responses noted by Casey et al. as data shows that only eNOS releases NO^• across the entire oxygen gradient from normoxia to complete anoxia (Mikula et al. 2009). Notwithstanding, it is important to note that the forearm and the leg vasculature show some small differences in their responses to vasodilators so the extent to which findings reported with forearm models can be extrapolated to vascular beds of other skeletal muscles requires caution. There may also be a large individual heterogeneity in this response.

Interpretation and implications

Whilst the study of Casey et al. (2010) investigated an isolated forearm skeletal muscle and resistance vessel bed the findings have extended our knowledge in the area of regulation of oxygen delivery in exercising and hypoxic tissue. Great interest has been placed on the potential therapeutic role of nitrite as a bioactive agent targeting hypoxic vessels. Important clinical findings such as blunted hypoxic pulmonary vasoconstriction (Ingram et al. 2010) and hypoxic augmentation of systemic venodilatation (Maher et al. 2008) have been noted in response to exogenous nitrite. However, it looks as though, in the face of an ablated endothelial regulation of NO^• release, the erythrocytic mechanism is incapable of stepping up NO^• production in healthy, young, exercising subjects. A point to note here is that the nitrite reductase activity of haemoglobin is arterial saturation () dependent, being maximal at around 50% (Gladwin, 2008). Therefore, in the study of Casey et al. (2010) the mean capillary may have been above the optimal hypoxic milieu to utilise the full function of the haemoglobin mechanism.

Clearly, whether the dominance of endothelial NO^• in hypoxic/ischaemic conditions can be recapitulated in ageing and disease cohorts is of clinical importance. Along these lines, future studies investigating endothelial NO^• donors such as l-arginine infusion during hypoxia would prove beneficial in fully teasing out the limitations of endothelial control of NO^• release. Moreover, a further interesting scenario that requires investigation is exogenous nitrite infusion in the presence of NOS blockade during hypoxia. This would completely isolate haemoglobin from endothelial pathways and may go some way in explaining the augmented vasodilatation effects of nitrite in hypoxia at rest (Maher et al. 2008). The role of the vasoconstrictor response to hypoxia cannot be overlooked; it may be that there is a down-regulation of receptors for metabolites such as angiotensin-II and endothelin-I favouring a net vasodilatation. Finally, our laboratory is at present engaged with research into oxidative–nitrative stress, hypoxia and vascular function. The stimulation of free radicals by hypoxic inspiration can inactivate blood-borne NO^• via rapid oxidation (k ≈ 10⁹m s⁻¹ between lipid-derived alkoxyl radical and NO^•) to yield the peroxynitrite anion [ONOO⁻], which has profound consequences on vascular tone (Bailey et al. 2009).

The exact candidate(s) for stimulation of endothelial NO^• in hypoxia remain to be determined and it is likely that a complex multifaceted system exists in a redox environment to defend the homeostatic O₂ gradient to active muscle. Thus, the study by Casey et al. (2010) has provided a key piece of the puzzle in trying to dissect out the origin of up-regulated NO^• production in hypoxic (exercising) tissue.