Nitrite regulates hypoxic vasodilation via myoglobin-dependent nitric oxide generation

Abstract

Background: Hypoxic vasodilation is a physiological response to low oxygen tension that increases blood supply to match metabolic demands. Although this response has been characterized for >100 years, the underlying hypoxic sensing and effector signaling mechanisms remain uncertain. We have shown that deoxygenated myoglobin in the heart can reduce nitrite to nitric oxide (NO·) and thereby contribute to cardiomyocyte NO· signaling during ischemia. On the basis of recent observations that myoglobin is expressed in the vasculature of hypoxia-tolerant fish, we hypothesized that endogenous nitrite may contribute to physiological hypoxic vasodilation via reactions with vascular myoglobin to form NO·.

Methods and results: We show in the present study that myoglobin is expressed in vascular smooth muscle and contributes significantly to nitrite-dependent hypoxic vasodilation in vivo and ex vivo. The generation of NO· from nitrite reduction by deoxygenated myoglobin activates canonical soluble guanylate cyclase/cGMP signaling pathways. In vivo and ex vivo vasodilation responses, the reduction of nitrite to NO·, and the subsequent signal transduction mechanisms were all significantly impaired in mice without myoglobin. Hypoxic vasodilation studies in myoglobin and endothelial and inducible NO synthase knockout models suggest that only myoglobin contributes to systemic hypoxic vasodilatory responses in mice.

Conclusions: Endogenous nitrite is a physiological effector of hypoxic vasodilation. Its reduction to NO· via the heme globin myoglobin enhances blood flow and matches O(2) supply to increased metabolic demands under hypoxic conditions.

Figure 1

Cyto-localization and myoglobin (Mb)-dependent nitrite reduction. A and B, Cyto-localization of Mb transcripts in the aorta of wild-type (A, Mb^+/+) and Mb deficient mice (B, Mb^−/−) by RNA-in situ hybridization. Using the anti-sense Mb riboprobe, smooth muscle cells exhibited a focal cytoplasmic signal (arrows) whilst the endothelium was negative. For both A and B Papanicolaou's haematoxylin counterstain was used. C and D, Immunohistology, using an antibody directed against the mouse Mb protein. While the Mb^+/+ aorta contains Mb in all smooth muscle cells (C), there is no evidence of Mb expression in Mb^−/− mice (D). E, Presence of Mb protein was confirmed using Western blotting. Mb protein and Mb^+/+ hearts served as positive controls. F through H, Nitrite-reductase activity of aortas from Mb^+/+ and Mb^−/− mice. F, Representative traces showing a decreased nitric oxide (NO^˙) formation in Mb^−/− compared to Mb^+/+ aortic tissue. G, Quantitative analysis reveals a significant difference between Mb^+/+ and Mb^−/− mice (mean±SD, *P<0.05 comparing Mb^+/+ and Mb^−/− mice, n=6–7). Inhibition of xanthine oxidoreductase (allopurinol + diphenyliodonium [DPI]) or blocking of mitochondrial respiratory chain (myxothiazol) did not significantly change NO^˙ release in either Mb^+/+ and Mb^−/− mice, while pre-incubation with ferricyanide to oxidize all cellular heme proteins significantly decreased NO^˙ generation in Mb^+/+ mice (mean±SD, ^#P<0.05 compared to untreated control). H, Addition of 20 μM hemoglobin (Hb) did not significantly change the rate of NO^˙ formation; control experiments using metHb. A statistical analysis of the Mb^−/− approaches (red columns) revealed a small but significant increase of nitrite reduction in Mb^−/− tissue after incubation with Hb.

Figure 2

Basal hypoxic vasodilation in vivo depends on nitrite reduction via myoglobin (Mb). A, Experimental design. Mice were mechanically ventilated and a Millar catheter was inserted into the right carotid artery to record systolic (P_sys) and diastolic (P_dias) pressure continuously. Following a normoxic equilibration period, the ventilation gas mixture was changed to 10% O₂/90% N₂ (induction of hypoxia). B and C, absolute and relative changes of hemodynamics, respectively, following the induction of hypoxia showing a greater decrease of pressures in wild-type (Mb^+/+) vs. Mb deficient (Mb^−/−) mice (asterisks and bars indicate time points and intervals with P<0.05 with n=7 and 6 respectively; values are means±SEM).

Figure 3

Nitrite-induced myoglobin (Mb)-dependent hypoxia vasodilation relies on NO^˙/sGC/cGMP signaling pathway and is independent of NO synthases. A compares the concentrations of RSNO levels in aortic tissue of wild-type (Mb^+/+) with Mb deficient (Mb^−/−) mice (mean±SD, n=3, *P<0.05), while B shows the same comparison for plasma and aortic tissue cGMP levels (mean±SD, n=3, *P<0.05). C shows the dependence of hypoxic vasodilation upon NO^˙/sGC/cGMP signaling in isolated aortic rings. Selective scavenging of NO^˙ with cPTIO or blocking of sGC via ODQ nearly abolished the vasodilatory response in vessel equilibrated under hypoxia and challenged with 10 μM nitrite. (mean±SD, *P<0.05 compared to controls, n=5). D, Formation of NO^˙ from nitrite was significantly higher in aortas of Mb^+/+ mice as detected by EPR spectroscopy. The displayed signals are representative of three independent experiments. Controls showed that spin trap (Fe-(DETC)₂) incubated with 1 mM [¹⁵N]nitrite and spin trap alone displayed no EPR signal. E, Quantitative analysis of these data revealed significantly higher amplitudes in Mb^+/+ aortas compared to Mb^−/− vessels (mean±SD, n=3, *P=0.018). F, In vivo vasodilation is sustained in mice lacking endothelial or inducible NO synthase (eNOS/iNOS). eNOS and iNOS are major alternative sources for NO^˙ under normoxia but ablation of either gene failed to affect the vasodilatory responses in vivo. Thus, hypoxically-induced vasodilation following the schema in Figure 2A was not reduced in eNOS^−/− (n=6) and iNOS^−/− (n=5) compared to the corresponding wild-type (C57BL/6) mice (n=5). Under normoxia the baseline pressures remained stable throughout (Figure 2 of the online-only Data Supplement), (values are means±SEM).

Figure 4

Nitrite-evoked vasodilation under hypoxia is dose dependent and reduced under myoglobin (Mb) deficiency. A, Experimental design and (B and C) relative effects of 16.7 and 1.67 μmol kg⁻¹ exogenous nitrite, respectively, upon hemodynamics under hypoxic ventilation. The relative decrease in P_sys and P_dias in wild-type (Mb^+/+) was significantly higher than in Mb deficient (Mb^−/−) mice (values are means±SEM, n=5, *P<0.05).

Figure 5

Formation of nitrosyl-myoglobin (MbNO) as an indirect marker for the formation of NO^˙ and independence of other nitrite-reductases. A, Exogenously applied nitrite was converted to NO^˙ and this nitrosylated Mb. Incubation of [¹⁵N]-labeled nitrite led to the formation of Mb[¹⁵N]NO as detected by EPR spectroscopy (deoxygenated Mb solution as authentic control). B, Nitrite reduction to vasodilatory NO^˙ was independent of xanthine oxidoreductase (inhibited by 100 μM allopurinol, 10 μM diphenyliodonium [DPI]), aldehyde oxidase (50 nM raloxifen) or mechanisms located in the endothelial layer (Endothel.) Compared to untreated controls, no significant decrease in vasorelaxation was detected whilst a significant difference between wild-type (Mb^+/+) and Mb deficient (Mb^−/−) aortic rings remained detectable (means±SD, n=3, *P<0.05, **P<0.01).

Figure 6

Cardiac function does not contribute to the decrease in blood pressure under hypoxia. A, Cardiac functions upon the induction of hypoxia as measured by an indwelling left ventricular pressure volume catheter. Either parameter shows a small increase which is incompatible with a contribution of cardiac function on hypoxic vasodilation (means±SEM). B, Wild-type (Mb^+/+) and myoglobin deficient (Mb^−/−) mice were anaesthetized and tracheally intubated. After stabilization, mice were challenged with hypoxia (10% O₂/90% N₂ – analogous to our in vivo protocol in Figure 2A). Chemiluminescence and HPLC were used to determine the plasma levels of nitrite and nitroso compounds (RNO). The latter comprises S-nitroso compounds (RSNO) and the remainder of bound NO^˙ (RXNO, e.g., N-nitroso compounds). No significant differences were measured between the two strains for either compound (means±SEM).