| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1Departments of Anesthesiology and Medicine, Pulmonary and Critical Care, University of Washington Veterans Affairs Puget Sound Health Care Center, Seattle, Washington 98104-2499; 2Department of Anesthesiology, College of Medicine, University of Ulsan, Asan Medical Center, Kangnung 682-714, Korea; 3Department of Anesthesiology, Myongji Hospital, Kwandong University, Kwandong 412-270, Korea; and 4Critical Care Medicine Department, Clinical Center, and Laboratory of Chemical Biology, National Institute of Diabetes and Digestive and Kidney Diseases, Bethesda, Maryland 20892
Submitted 29 March 2004 ; accepted in final form 26 July 2004
 |
ABSTRACT |
|---|
 TOP ABSTRACT MATERIALS AND METHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
nitric oxide; hypoxic pulmonary vasoconstriction; erythrocytes
-cysteine 93 residue, NO can bind reversibly with Hb to form SNO-Hb (12). The reverse reaction is accelerated by low-molecular-weight thiols, such as GSH (5, 6, 12, 26). It has also been proposed that release of NO from SNO-Hb is under allosteric control and is promoted by deoxygenation of Hb (R
T structural transitions) (1, 12, 23). Consistent with this allosteric behavior, the T-state SNO-Hb is proposed to interact with the band 3 protein (anion exchanger-1; AE-1), transferring NO from the -cysteine93 residue extracellularly (18). These properties would confer on Hb and red blood cells (RBCs) an active role in hypoxic vasomoter regulation, whereby NO is carried by RBCs from lung to peripheral tissues and then released in proportion to local tissue oxygen demand and concentration, thus increasing blood flow to vascular beds that need it most (12, 15, 23). However, this allosteric mechanism has been extensively challenged (3, 5–7, 9, 27). We (5, 6) and others (9, 15, 17) have reported that SNO-Hb degrades in the presence of millimolar glutathione (5 mM in the erythrocyte) independent of PO2. Furthermore, the apparent allosteric vasodilatory effects of SNO-Hb may be secondary to simple hypoxic potentiation of all NO donors (not requiring oxygen-dependent allosteric structural transitions intrinsic to the Hb molecule) (3).
To critically evaluate the allosteric behavior of SNO-Hb, we have developed a methodology to synthesize relatively high concentrations of intraerythrocytic SNO-Hb. These erythrocytes have normal oxygen affinity, contain the physiological complement of the intact erythrocyte (2,3-diphosphoglycerate, AE-1, glutathione, etc.), have relatively little metHb, and contain concentrations of SNO-Hb up to 1,000 times the normal blood levels (9). As such, these cells provide for a more physiological assessment of the properties of SNO-Hb than that of pure Hb preparations with added glutathione and organic phosphates to regulate Hb oxygen affinity.
The pulmonary circulation is a uniquely suited model for the study of oxygen-regulated vasodilators for several reasons. First, the response of the pulmonary circulation to hypoxia is vasoconstriction, in contrast to the systemic circulation, which vasodilates. This provides a strong physiological background signal for the study of vasodilator effects. Second, although SNO-Hb would not be expected in vivo to release NO in the pulmonary circulation because the diffusion of oxygen is from the alveolar gas into blood, in the isolated lung rapid reduction of alveolar PO2 by ventilation with hypoxic gas reverses the flux of oxygen and thus provides a model for systemic capillary O2 exchange in hypoxic tissues. Thus, if release of SNO or NO from SNO-Hb is allosterically facilitated by deoxygenation, the induction of hypoxia in the isolated lung will result in release of vasodilator substances that will oppose intrinsic hypoxic pulmonary vasoconstriction. In the present study, we studied the effects of human erythrocytes enriched with supraphysiological concentrations of SNO-Hb on pulmonary vasoconstriction induced either pharmacologically or by hypoxia in an ex vivo model to determine the following: 1) whether SNO-Hb within the erythrocyte has physiologically relevant vasodilatory activity, 2) the role of heme deoxygenation on release of SNO from Hb and resulting vasoactivity, and 3) the role of AE-1 on NO export from these erythrocytes.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
Reagents. All reagents were purchased from Sigma (St. Louis, MO), unless otherwise noted.
RBC preparation. RBCs containing SNO-Hb (SNO-RBCs) were synthesized as previously described (9). Briefly, human blood was incubated at 4°C with excess S-nitrosocysteine (Cys-NO) for 3–4 h and then washed five times to remove excess Cys-NO. The consolidated RBC pellet was stored at 4°C until just before addition to the perfusate. Samples of supernatant from the final RBC wash and from the RBC pellet were flash-frozen for later determination of the concentration of residual Cys-NO (wash) and SNO-Hb (pellet). MetHb concentration was determined spectrophotometrically in a fresh RBC sample (25). Control RBCs were prepared and stored in identical fashion to that described above, in the absence of Cys-NO addition.
Hb preparation. OxyHb was prepared from fresh human blood as previously described (5) and frozen at –80°C for later use. Spectrophotometrically derived metHb concentration was <1% before addition to the lung perfusate (25).
Measurement of SNO-Hb and perfusate SNO. Extraerythrocytic SNO concentration and the concentration of SNO-Hb in erythrocytes was determined by tri-iodide-based reductive chemiluminescence (9). Perfusate samples were collected at the intervals described below, and immediately centrifuged at 2,500 rpm for 5 min to separate cells from supernatant. For the majority of samples, the supernatant was treated with N-ethylmaleimide (NEM) and diethylenetriamine pentaacetic acid water and flash frozen, whereas the RBC pellet was lysed 1:10 in a SNO-Hb "stabilization solution" of PBS containing 1% Nonidet P-40 (to solubilize membranes), 8 mM NEM (to bind free thiol and prevent artefactual SNO), 0.1 mM diethylenetriamine pentaacetic acid (to chelate trace copper), and 4 mM ferricyanide and cyanide (to stabilize SNO-Hb and prevent artifactual ex vivo iron nitrosylation during processing) (9, 28). For the last eight experiments involving DIDS (see below), the SNO-stabilization solution was not used. The Hb samples were desalted across a 9.5-ml bed volume Sephadex G25 column to eliminate nitrite and excess reagents and partially purify Hb (99% Hb preparation). The Hb fraction was quantified by the method of Drabkin, and Hb fractions reacted with and without mercuric chloride (1:5 HgCl2-to-heme ratio used to differentiate S-nitrosothiol, which is mercury labile vs. iron-nitrosyl, which is mercury stable) and then in 0.1 mol/l HCl-0.5% sulfanilamide (to eliminate residual nitrite) (14). The samples were then injected into a solution of tri-iodide in line with a chemiluminescent NO analyzer (model 280, Sievers; Boulder, CO). The mercury-stable peak represents iron-nitrosyl-Hb. This assay is sensitive and specific for both SNO-Hb and iron-nitrosyl-Hb to 5 nM in whole blood (0.00005% SNO per heme) (9).
To determine the relative content of SNO-Hb versus other SNO erythrocyte proteins, detection of SNO proteins in RBCs was carried out using a specific SNO biotinylation assay (11). SNO-RBCs (0.1 ml) were transferred into a tube containing SNO-Hb stabilization solution (28). Ten microliters of the reaction mixture were incubated with blocking buffer (250 mM HEPES, pH 7.7, 1 mM EDTA, 0.1 mM neocuproine, 20 mM NEM, 2% SDS, and 2 mM potassium ferricyanide) (90 µl) for 30 min. After passage through microspin G25 columns, 10 µl of 4 mM biotin-N-[6(biotinamido)hexyl]-3'-(2'-pyridylthio)propionamide were added (final concentration is 0.4 mM), followed by 2 µl of Na ascorbate (50 mM). After biotinylation was completed, 10 µl of solution were transferred into a tube containing 190 µl of nonreducing SDS sample buffer (x20 dilution). Ten microliters of the reaction mixture were electrophoretically separated on 4–20% nonreducing SDS-PAGE. Proteins were transferred onto nitrocellulose membrane. Immunoblot analysis was carried out with the use of peroxidase conjugated anti-biotin antibodies (Jackson ImmunoResearch Laboratories; West Grove, PA) and the percentage S-nitrosated Hb (16-kDa band) was calculated with the use of densitometry. The percentages of metHb and oxyHb were measured in iced perfusate samples using a commercial CO-oximeter.
Experimental preparation.
The isolated, perfused, lung model is similar to that described by others (22). Briefly, adult Sprague-Dawley rats weighing 300–500 g were anesthetized with intraperitoneal pentobarbital, a tracheotomy was established, and mechanical ventilation was initiated with a mixture of 21% O2-6% CO2-balance N2, using a respiratory rate of 
30 breaths/min and a peak airway pressure of 10 cmH2O. Heparin (300 units) was administered by intracardiac injection, the pulmonary artery and left atrium were cannulated in situ, and the lungs were flushed of blood by slow perfusion with a Masterflex pump (Cole-Palmer; Barrington, IL). After being flushed, a recirculating circuit was established with a total circuit volume of 15 ml at a constant flow of 15 ml/min. The heart and lungs remained in situ throughout the experiments, and were covered with plastic wrap to maintain temperature and humidity. The perfusate consisted of buffered crystalloid solution composed of (in mM) 11 D-glucose, 1.2 MgSO4, 1.2 KH2PO4, 4 KCl, 118 NaCl, 2.5 CaCl2, and 25 NaHCO3, containing 4% Ficoll as a colloid. Meclofenamate (3.1 µM) was added to the perfusate to prevent any effects of eicosanoids on vascular tone or the stability of the preparation. Perfusate pH was maintained at 7.35–7.45 by addition of NaHCO3 as needed. Left atrial pressure was maintained constant at 2 Torr with a screw clamp on the outflow circuit, and perfusate temperature was maintained at 38 ± 0.5°C. Preparations were allowed to stabilize for 20 min before any experimental interventions.
Pulmonary artery pressure (PAP) and left atrial pressure were measured via a small-bore tubing that was positioned internally at the ostia of the perfusion cannulae (4). Pressure data were continuously recorded using an analog-to-digital converter, data-acquisition software (Strawberry Tree; Sunnyvale, CA), and a personal computer (Macintosh; Cupertino, CA). Perfusate PO2, PCO2, and pH were measured using a conventional blood-gas machine (Radiometer; Copenhagen, Denmark).
At the conclusion of experiments utilizing RBCs, perfusate samples were obtained for measurement of free Hb, which was then measured using a commercially available colorimetric benzidine oxidation assay (Sigma Diagnostics). Absorbance at a wavelength of 600 nm is linear between Hb concentrations of 1–10 µM with this assay.
Experimental protocol: effect of RBCs containing SNO-Hb on PAP under normoxic and hypoxic conditions. After an initial period of stabilization as described above, GSH was added to the perfusate to achieve a concentration of 500 µmol/l. The lungs were then ventilated with anoxic, normocarbic gases (0 inspired O2 fraction-0.06 inspired CO2 fraction-balance N2); after 5 min, either SNO-RBCs or control RBCs that had been warmed for 5 min to near 37°C were added to the perfusate in an amount sufficient to achieve an Hct of 10–15%. Anoxic ventilation was continued for another 10 min (total of 15 min), and the lungs were then returned to normoxic ventilation for a final 10 min. The lungs were then flushed with fresh buffer until the perfusate was clear, and the protocol repeated using the alternate RBC type (control or SNO-RBCs).
In a separate series of experiments, a protocol identical to the above was followed; however, RBCs were added to the perfusate during an initial period of normoxic ventilation. After 10 min of normoxic perfusion, anoxic ventilation was instituted and the lungs were perfused for an additional 10 min under hypoxic conditions.
Initially, the SNO-Hb content of RBCs was measured in the cold pellet and in perfusate samples taken at 5, 10, and 20 min of perfusion. After preliminary analysis revealed rapid loss of SNO from the cold pellet to 5 min of perfusion, samples were also taken from the warm pellet and at 30 s of perfusion to allow better characterization of SNO loss.
Experimental protocol: effects of SNO-RBCs on pulmonary hypertension induced by U-46619. After surgical preparation and stabilization and during normoxic ventilation, acute pulmonary hypertension was produced by continuous infusion of U-46619 (U-4), a thromboxane mimetic, at 0.015–0.12 ng/min. Once a stable pulmonary artery pressure (15–20 mmHg) was obtained, the U-4 infusion was reduced by 50%, and PAP was followed for an additional 10 min to ensure stability. Reversibility of pulmonary hypertension was established by administration of a known nitrosothiol vasodilator, S-nitrosoglutathione (GSNO), at a starting concentration of 0.5 µmol/l. When PAP had recovered to previously hypertensive values for at least 10 min, either SNO-RBCs or control RBCs were added to the perfusate over a 5-min period to obtain a perfusate Hct of 10–15%, and PAP was recorded at 5-min intervals for 10 min. GSNO was then again added to the perfusate at increasing doses at 5-min intervals, and PAP recorded.
Experimental protocol: effect of GSNO on hypoxic pulmonary vasoconstriction. To compare GSNO-induced vasodilation during hypoxia with vasodilation during normoxia and U-4 administration, hypoxic pulmonary vasoconstriction was provoked in buffer-perfused lungs by ventilation with anoxic gas for 5 min. This protocol was then repeated after sequential administration of 0.1, 0.5, and 5 µM GSNO to the perfusate. The change in PAP provoked by hypoxia versus U-4 administration was then compared at baseline and GSNO concentrations of 0.5 and 5 µM using repeated-measures ANOVA.
Experimental protocol: role of AE-1 on SNO-RBC-induced vasodilation.
A protocol identical to that described above was followed with the exception that either DIDS or dinitrostilbene disulfonate (DNDS), inhibitors of the RBC membrane protein AE-1, were added to SNO-RBCs before the last wash, and to the lung perfusate 5 min before RBC addition at a concentration of 100 µM DNDS or 100 and 500 µM DIDS. RBCs were added during anoxic ventilation, as described above. Because the results of these initial experiments were equivocal due to a weak pressor response during hypoxia, an additional series of experiments were conducted after addition of 100 µM N
-nitro-L-arginine methyl ester to the perfusate. In these experiments, the increase in PAP during anoxic ventilation was tested in the presence of control RBCs or SNO-RBCs, with both groups of cells incubated with 100 µM DIDS before the addition to the perfusate, plus 100 µM DIDS in the perfusate. Finally, to determine whether any effects of DIDS or DNDS on PAP during hypoxic ventilation were specific to effects on the RBC membrane, an additional series of experiments was performed using free oxyHb (100 µM) as a control.
Statistical analysis. Statistical analysis was performed with the use of StatView software (Abacus Concepts; Berkeley, CA). Data are presented as means ± SE. Interventions performed within individual experiments were compared using a paired t-test. Group comparisons were performed using one-way ANOVA at individual time points, or repeated-measures ANOVA for interaction effects. The Bonferroni-Dunn correction for multiple comparisons was performed where appropriate. A P < 0.05 was accepted as statistically significant.
|
RESULTS |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
1.2%), with minimal iron-nitrosyl(heme)Hb formation. MetHb concentration ranged between 0.6 and 13.6% (mean 6.1 ± 0.6%) in the cold RBC pellet, with an r value of 0.72 (P = 0.017) for the correlation between synthesized SNO-Hb and metHb. The SNO concentration in the final RBC buffer wash was 110 ± 36 nM.
Biotinylation confirmed that 93% of the SNO protein within RBCs was Hb (Fig. 1). Other experiments demonstrated specific modification of the -subunit (data not shown).
|
PAP = 2.5 ± 1 vs. 1.8 ± 0.6 Torr, control vs. SNO-RBC groups). U-4 infusion during normoxic ventilation in buffer-perfused lungs resulted in an 5 Torr increase in PAP, which did not differ between groups (PAP = 5.3 ± 0.4 vs. 4.8 ± 0.3 Torr, control vs. SNO-RBC groups). The remainder of the data is expressed as the change in PAP (PAP) from the normoxic, normotensive pressure during buffer perfusion. The addition of both control and SNO-RBCs to the perfusate during normoxic, ventilation resulted in a small increase in PAP, without a significant difference between groups (Fig. 2A). During normoxic conditions and U-4 administration, PAP increased further with addition of control RBCs to the perfusate (P < 0.05), but it did not increase compared with buffer plus U-4 after addition of SNO-RBCs, consistent with a dilatory effect of SNO-RBCs under normoxic conditions. The addition of RBCs to the lung perfusate during anoxic ventilation resulted in an increase in PAP. However, PAP increased more when control RBCs were added versus after the addition of SNO-RBCs. However, initiation of anoxic ventilation after 10 min of normoxic ventilation (delayed hypoxia) resulted in an increase in PAP in both groups, with no significant difference between groups. The lack of relative vasodilation by SNO-RBCs at this point was likely due a depletion of SNO-Hb and perfusate SNO due to rewarming and circulation (data discussed below). In support of this, the inhibitory effect of SNO-RBCs on hypoxic pulmonary vasoconstriction was dose dependent. When the SNO content of RBCs was arbitrarily divided into those cells containing less than or >1% SNO/heme, RBCs containing a low concentration of SNO (0.48 ± 0.9%) reduced hypoxic PAP compared with control RBCs only slightly, whereas RBCs containing a high concentration of SNO (1.88 ± 0.44%) dramatically reduced PAP (Fig. 2B).
|
7 Torr increase in PAP from the untreated baseline (Fig. 3). GSNO reduced PAP in a dose-dependent fashion during both normoxic (U-4 induced) and hypoxic pulmonary vasoconstriction, but this effect was much greater during hypoxia (Fig. 3) (P < 0.03 for GSNO-group interaction). In two experiments, 0.1 µM GSNO did not affect PAP during hypoxia (data not shown). GSNO continued to exhibit dose-dependent vasodilation in U-4-treated lungs at does up to 200 µM. Higher concentrations of GSNO were not administered during hypoxia due to complete inhibition of vasoconstriction by 5 µM GSNO in four of five experiments.
|
After the large drop in SNO/heme was noted in hypoxia experiments, additional samples were taken from the warm RBC pellet (after 5 min of warming) and from the perfusate 30 s after the addition of cells to better characterize the loss of SNO. Figure 4 illustrates the fall in SNO/heme associated with RBC warming, and addition to and dilution by the perfusate. There was considerable loss of SNO during rewarming, and the average SNO/heme in the warm pellet was 0.84 ± 0.15% and 0.95 ± 0.15% in the hypoxic and normoxic groups, respectively, translating to immediately available concentrations of SNO-Hb in the perfusate of 25 µM for both groups. An additional 40% relative reduction in SNO/heme occurred within 30 s of RBC addition to the perfusate, with a slower rate of loss over the next several min. Hypoxia exerted no influence on the rate of SNO loss from Hb. The extraerythrocytic SNO concentration (from supernatant) was 200 nM 5 min after addition of RBCs to the perfusate, and there were no significant differences in these concentrations between normoxic and hypoxic conditions at any time (data shown at 10 min; Fig. 5). Likewise, perfusate nitrite did not differ significantly between groups (Fig. 5). Extraerythrocytic SNO fell gradually over 10 min of perfusion during both hypoxia and normoxia. Perfusate nitrite concentration was 500 nM.
|
|
Role of AE-1 on inhibition of hypoxic vasoconstriction and loss of SNO by SNO-RBCs.
Incubation of SNO-RBCs with DIDS or DNDS had no significant effect on PAP during anoxic ventilation compared with untreated SNO-RBCs (data not shown). Hypoxic vasoconstriction was weak in these studies compared with previous experiments using control cells without SNO. However, in lungs treated with N-nitro-L-arginine methyl ester to maximize hypoxic vasoconstriction, SNO-RBCs continued to inhibit hypoxic vasoconstriction compared with control RBCs despite treatment with DIDS (P < 0.05 between groups) (Fig. 6A). DNDS had no significant effect on the rate of loss of SNO from Hb before or after the addition of SNO-RBCs to the perfusate (data not shown; P = 0.93, repeated-measures ANOVA). Likewise, DIDS had no effect on the rate of loss of SNO from Hb (Fig. 6B) (P = 0.68, repeated-measures ANOVA). This was true for DIDS concentrations of 100 µM (n = 5) and 500 µM (n = 3) (data in Fig. 6B represent experiments using both concentrations). Neither DNDS nor DIDS had any effect on normoxic or hypoxic PAP when added to perfusate containing free Hb (n = 3 in each group, data not shown).
|
|
DISCUSSION |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
Pulmonary vasodilation by SNO-RBCs. Our studies and those of others have shown that free Hb and RBCs augment hypoxic pulmonary vasoconstriction by inactivating nitric oxide. Free SNO-Hb augments hypoxic pulmonary vasoconstriction to the same degree as non-S-nitrosated oxyHb, as oxidative and additive reactions of NO with heme overwhelm any potential SNO-donating effects (5, 6). In this study, RBCs containing SNO-Hb continue to augment hypoxic pulmonary vasoconstriction, but this effect is much reduced compared with control RBCs not containing SNO (Fig. 2, A and B). This appears to occur as a result of rapid, temperature-dependent loss of SNO from Hb and interaction with the pulmonary vasculature in the first minutes after addition to the circulating perfusate, in combination with sequestration of Hb to the RBC interior and away from extraerythrocytic SNO. This is consistent with the function of the RBC membrane as a diffusion barrier that limits NO inactivation (10, 13, 20, 24).
Vasodilation by RBCs containing SNO-Hb is dose dependent, with higher concentrations nearly ablating the increase in PAP with hypoxia and with lower concentrations mildly reducing (Fig. 2B) or having no effect on PAP (Fig. 2A, delayed hypoxia). However, even the lowest concentrations of SNO-Hb observed in this study remain far in excess of those measured physiologically. For example, by extrapolation from the SNO/heme concentration in warmed RBCs just before addition to the perfusate, the immediately available perfusate SNO concentration contributed by Hb was 25 µM. This SNO was rapidly lost from the RBC, however, and after 15 min of perfusion the effective SNO-Hb concentration was reduced to <10 µM. At this point, SNO-RBCs no longer reduced hypoxic pulmonary vasoconstriction compared with control RBCs (Fig. 2A, delayed hypoxia). In contrast, rat arterial blood SNO-Hb concentrations of only 300 nM have been reported (12, 19). More relevant with regard to translation to human physiology, we have measured SNO-Hb concentrations of 45–70 nM in human blood (8), and Rassaf et al. (19) were unable to detect SNO-Hb down to a detection limit of 10–20 nM. Given these low in vivo SNO-Hb concentrations, it seems highly unlikely that SNO-Hb plays an important role in the regulation of hypoxic pulmonary vasoconstriction and pulmonary blood flow under normal physiological conditions, although this cannot be said with certainty given that our studies were performed in isolated perfused lungs.
The SNO concentration in the last RBC preparatory wash, which represents residual Cys-NO, was slightly >100 nM. The final concentration of Cys-NO in the lung perfusate would represent less than one-third of this value, given that 3–5 ml of packed RBCs were added to buffer perfusate to create a total volume of 15 ml. Although Cys-NO undoubtedly contributed in small part to the total perfusate SNO concentration, it is unlikely that the contribution of Cys-NO alone would be sufficient to cause vasodilation in our model.
Consistent with previous studies, hypoxia is not necessary for RBCs containing SNO-Hb to either release SNO or to exert a vasodilatory effect (2, 3). Under normoxic, normotensive conditions, where the increase in PAP in RBC versus buffer-perfused lungs is largely related to viscosity, the effect on PAP did not differ between control and RBCs containing SNO-Hb (Fig. 2A). However, in lungs treated with the thromboxane analog U-4, control RBCs resulted in a further increase in PAP, whereas SNO-RBCs did not (Fig. 2A), implying vasodilation by the latter. In addition, SNO-Hb concentrations fall at the same rate under both normoxic and hypoxic conditions (Fig. 4). In aggregate, these data are not consistent with an oxygen-linked allosteric delivery of NO from SNO-Hb.
In our model, SNO-RBCs reduced vasoconstriction by 50% in both hypoxic and normoxic conditions (Fig. 2A). It would not be surprising if SNO-RBCs had a greater effect during hypoxia, given the relative increase in vasodilator potency of SNO-Hb, GSNO, nitroxyl anion, and other NO donors in system vascular smooth muscle under hypoxic conditions (3). We have confirmed that this effect is also present in the pulmonary circulation by showing greater GSNO-induced vasodilation during hypoxic versus normoxic conditions (Fig. 3). These critical control experiments help explain the observation that the rate of SNO-Hb decay and SNO export is independent of PO2, whereas the vasodilating effect may increase as oxygen tension decreases. The present study, by utilizing intact RBCs, also obviates concerns that the lack of allosterically linked vasodilation by free SNO-Hb shown by Crawford et al. (3, 16) was due to Hb dimer formation. Thus although RBCs containing high concentrations of SNO-Hb may be capable of exerting oxygen-dependent vasodilatory activity, this effect is unrelated to Hb allostery.
Because of the rapid inactivation of NO by free Hb, hemolysis in our perfusion system could certainly confound our findings. However, we found that free Hb concentrations in the perfusate at the conclusion of experiments were 5 µM in both control and SNO-RBC groups. We have previously shown that this concentration of free Hb is too low to cause significant vascular effects in isolated rat lungs (6).
Role of AE-1 in export of SNO from RBCs.
Pawloski et al. (18) have suggested that export of NO from SNO-Hb containing RBCs is facilitated by the membrane ion transport protein AE-1. Intracellular Hb is associated closely with AE-1, and this close association may protect SNO/NO from reduction and dioxygenation by heme when it is released from the -cysteine 93 residue. However, our data suggest that this mechanism is relatively unimportant. The AE-1 inhibitors DIDS and DNDS at fully inhibiting concentrations for anion exchange failed to prevent inhibition of hypoxic pulmonary vasoconstriction by SNO-RBCs, and they did not slow the rate of loss of SNO from RBCs (Fig. 6). These data suggest that additional mechanisms for the release of SNO from Hb and export from the RBC must be important. These may include reactions of SNO-Hb with small-molecular-weight thiols, such as GSH, which is present in high concentration within the RBC cytoplasm, to form GSNO (12, 17, 21), the nitroxyl anion (26), nitrite (2), or other yet-to-be-determined mechanisms.
Allosteric influences on SNO loss from Hb.
In this study, as in our two previous studies of SNO-Hb in isolated, perfused lungs, we found no evidence that deoxygenation influences the rate of SNO loss from Hb (Fig. 4) (5, 6). Indeed, synthesized SNO-Hb, even when encapsulated by the RBC membrane, is extremely labile and degrades rapidly at room temperature (Fig. 4). These findings are consistent with the in vitro findings of SNO-Hb biochemistry described by multiple investigators (9, 17, 26, 27). Although the present study was not designed to directly test the role of SNO-Hb in systemic circulatory physiology, our findings raise doubts about the importance of SNO-Hb as a delivery mechanism for NO. SNO-Hb-containing erythrocytes appear to be a relatively weak vasodilators in our system, likely due to the competing effects of NO oxidation by heme and NO release into the circulation from the -cysteine 93 residue, and it is unlikely that the low-circulating concentrations and the small concentration differences occurring with normal arterial-venous Hb O2 saturation changes found in most studies exert a significant physiological effect. SNO-RBCs are also unlikely to have major therapeutic potential due to the lability and short half-life of SNO-Hb. However, it is possible that in pathophysiological states, such as sepsis, where continuous NO production is high, SNO-Hb formation may be increased sufficiently that it could exert vasodilator effects. Further study is necessary to examine these possibilities.
In conclusion, RBCs containing high concentrations of SNO-Hb oppose vasoconstriction in a pulmonary circulation preconstricted pharmacologically or by hypoxia. However, the effects do not appear to be mediated by O2-linked Hb allosteric changes or involve export via membrane AE-1.
|
GRANTS |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
|
FOOTNOTES |
|---|
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
|
REFERENCES |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
-93 thiol of human hemoglobin during cycles of oxygenation and deoxygenation. Proc Natl Acad Sci USA 100: 11303–11308, 2003.This article has been cited by other articles:
|
|
 |
|
  P. Sonveaux, I. I. Lobysheva, O. Feron, and T. J. McMahon Transport and Peripheral Bioactivities of Nitrogen Oxides Carried by Red Blood Cell Hemoglobin: Role in Oxygen Delivery Physiology, April 1, 2007; 22(2): 97 - 112. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |
P < 0.0001 vs. buffer; 
P < 0.01 vs. buffer. B: SNO-RBCs were arbitrarily divided into those with a low baseline SNO content [SNO/heme <1% (0.48 ± 0.11%)] or a high baseline SNO content [SNO/heme >1% (1.88 ± 0.47%)] (n = 4–12 per group). *P < 0.05 vs. buffer; 
0.05 vs. control RBCs. SNO-RBCs reduced hypoxic pulmonary vasoconstriction despite treatment with DIDS. B: SNO-RBCs or SNO-RBCs incubated with DIDS (100–500 µmol/l) were added to the perfusate of lungs during anoxic ventilation. DIDS was also added to the perfusate at the concentrations above. SNO/heme was measured in the cold RBC pellet (baseline, –5 min), after rewarming to 37°C (time 0), and at 30 s, and 5 and 10 min after addition to buffer-perfused lungs ventilated with anoxic gas, and is reported as a percentage of the baseline (cold pellet) value. DIDS had no effect on the rate of loss of SNO from RBCs (n = 8).