Oxygen Regulation of Tumor Perfusion by S-Nitrosohemoglobin Reveals a Pressor Activity of Nitric Oxide
http://www.100md.com
循环研究杂志 2005年第5期
The Department of Radiation Oncology (P.S., A.M.K., S.A.S., R.A.R., L.I.C.-N., R.D.B., M.W.D.), Howard Hughes Medical Institute (J.S.S.) and Department of Medicine (J.R.P., T.J.M., J.S.S.), Duke University Medical Center, Durham, NC
Nicholas School of the Environment and Earth Sciences, Duke University Marine Laboratory
University of Puerto Rico NIH COBREII Protein Research Center, Beaufort, NC and Mayaguez PR00681 (J.B.)
Tumor Microcirculation Group, Gray Cancer Institute, Mount Vernon Hospital, Northwood, Middlesex, UK (G.M.T).
Abstract
In erythrocytes, S-nitrosohemoglobin (SNO-Hb) arises from S-nitrosylation of oxygenated hemoglobin (Hb). It has been shown that SNO-Hb behaves as a nitric oxide (NO) donor at low oxygen tensions. This property, in combination with oxygen transport capacity, suggests that SNO-Hb may have unique potential to reoxygenate hypoxic tissues. The present study was designed to test the idea that the allosteric properties of SNO-Hb could be manipulated to enhance oxygen delivery in a hypoxic tumor. Using Laser Doppler flowmetry, we showed that SNO-Hb infusion to animals breathing 21% O2 reduced tumor perfusion without affecting blood pressure and heart rate. Raising the pO2 (100% O2) slowed the release of NO bioactivity from SNO-Hb (ie, prolonged the plasma half-life of the SNO in Hb), preserved tumor perfusion, and raised the blood pressure. In contrast, native Hb reduced both tumor perfusion and heart rate independently of the oxygen concentration of the inhaled gas, and did not elicit hypertensive effects. Window chamber (to image tumor arteriolar reactivity in vivo) and hemodynamic measurements indicated that the preservation of tissue perfusion by micromolar concentrations of SNO-Hb is a composite effect created by reduced peripheral vascular resistance and direct inhibition of the baroreceptor reflex, leading to increased blood pressure. Overall, these results indicate that the properties of SNO-Hb are attributable to allosteric control of NO release by oxygen in central as well as peripheral issues.
Key Words: nitric oxide hemoglobin oxygen hemodynamics blood flow
Introduction
Hemoglobin (Hb) of red blood cells (RBC) is a tetrameric protein composed of 2 and 2 chains, each containing a heme prosthetic group. One and 1 chain is combined in stable dimers, and 2 dimers are more loosely associated to form tetramers. The physiological role of Hb depends on its ability to reversibly bind O2 at its heme iron centers. This transport capacity is governed by a cycle of allosteric transitions in which Hb assumes the R (relaxed, high O2 affinity) conformation to bind O2 in the lungs and, on partial deoxygenation, the T (tense, low O2 affinity) conformation to efficiently deliver O2 to peripheral tissues. This transition also controls the reactivity of 2 conserved cysteines on the chains (Cys93). Thiols of the cysteines can react with the potent vasodilator nitric oxide (NO) to form S-nitrosohemoglobin(SNO-Hb) in the R conformation, and preferentially unload SNO in the T conformation.1eC3 It has therefore been proposed that Hb would carry NO equivalents from the lungs to the periphery, thereby bringing tissue blood flow in line with oxygen demand. Although mechanisms are debated,4eC7 some consensus has been reached, namely that Hb can act as an oxygen sensor and oxygen-dependent transducer of NO bioactivity,7eC9 and the direct coupling between SNO and O2 content of Hb has been recently affirmed in intact human erythrocytes (Doctor et al, unpublished observations).
Outside the red blood cell (RBC), Hb has toxic effects that prevent its use as a blood substitute in humans.10 In plasma, Hb rapidly dissociates to dimers, leading to hemoglobinuria (renal injury) and extravasation (osmotic disorders).11 Several types of conjugated and intramolecularly cross-linked Hbs have been developed that reduce Hb dissociation, thereby lowering the risks for such toxicity.12 These modified Hbs also show prolonged vascular retention times, making them promising blood substitute candidates. However, clinical safety is still not achieved, mainly because modified Hbs retain the vasoconstrictive activity of cell-free Hb.13 Ferrous Hb and oxy-Hb can react with NO to produce iron-nitrosyl Hb, and oxidized Hb and nitrate ions (met-Hb reaction), respectively.14,15 The Hb concentrations required to exert such activity have not been carefully examined, and the extent to which vasodilation is involved (under more physiological circumstances, Hb can also react with NO or S-nitrosothiols to yield SNO-Hb)16 remains a matter of debate. Lower doses of Hb-based blood substitutes do not typically raise blood pressure.
To date, SNO-Hb, which combines the capacity to carry oxygen and the potential to act as a hypoxia-activated NO donor, has not yet been fully tested as a RBC substitute.17 In the present work, we hypothesized that S-nitrosylation of cell-free Hb would reduce the vasoconstrictive effects of tetrameric Hb. We also hypothesized that blood oxygenation, through allostery-regulated control of NO release, would be a key determinant of the biological responses to SNO-Hb. These hypotheses were investigated in vivo using a physiological model of vascular hypoxia (perfused tumor vessels), in which pO2 was modulated in situ by varying inspired O2. This prospective study offers a new perspective on the development of transfusion substitutes, and establishes that delivery of NO bioactivity by SNO-Hb is regulated by O2 saturation.
Materials and Methods
Rats and Tumors
Seventy-four female Fischer 344 rats (Charles River, Raleigh, NC), bearing or not bearing the syngeneic mammary adenocarcinoma R3230AC,18 were used in all experiments. Tumor pieces from donor rats were implanted in anesthetized animals (100 mg/kg ketamine and 10 mg/kg xylazine i.p.). Tumors were grown in the subcutis of the left lateral quadriceps muscle or, for window chamber experiments, between 2 fascial layers on the back of rats. Animals were then randomly assigned to a treatment group. Further interventions/observations were performed on anesthetized animals (50 mg/kg pentobarbital i.p.) maintained at 37°C on a temperature-controlled thermal blanket. Where indicated, some rats were fitted with a facemask for breathing 100% O2. All experiments were approved by Duke IACUC.
Drugs
SNO-Hb was synthesized through reaction of purified human hemoglobin A0 (Apex Bioscience, Durham, NC) with SNO-cysteine in nonacidic conditions, as previously reported19 (expanded method at http://www.circresaha.org). SNO-albumin was synthesized by the reaction of an equal volume of bovine serum albumin (0.2 mmol/L in HCl 0.1 mmol/L, 0.5 mmol/L EDTA) with 0.2 mmol/L NaNO2 for 30 minutes. All solutions (diluted at 200 eol/L in PBS pH 7.4, 0.5 mmol/L EDTA) were kept on ice or frozen and protected from light until administration. Less than 2-weeks old solutions were infused at a dose of 200 nmol/kg in 0.25 mL, immediately followed by a 0.25 mL delivery of saline. Assuming a plasma volume of 3 mL per 100 g in rats and no significant partition in blood cells, this dose would achieve a maximum plasma concentration of 6.5 eol/L for all drugs. To avoid volume-induced alterations of hemodynamics, solutions were infused at 0.5 mL/min from time 0.
Surgery
For i.v. infusions, the femoral artery and vein were cannulated and drugs were infused through the venous cannula. For i.a. infusions, the femoral and left carotid arteries were cannulated and drugs were infused through the carotid cannula. For window chamber experiments, the 2 window frames were surgically positioned in the dorsal skin flap, followed by tumor cell transplantation, as previously described.20 Following several days of tumor growth, direct visualization of tumor-feeding arterioles was performed using intravital microscopy.
Blood Gas and Hemoglobin Saturation Measurements in Tumor Microvessels
Blood gases and Hb oxygen saturation were measured using a 1640 oximeter coupled to a 482 cooximeter (Instrumentation Laboratories, Lexington, Mass) from 0.5 mL samples of femoral arterial and venous blood. For tumor microvessels in window chambers, Hb saturation was calculated from transmission optical measurements of vascular absorbance using a hyperspectral imaging technique.21 A liquid crystal tunable filter (CRI, Inc., Woburn, Mass) placed in front of a CCD camera (DVC Company, Austin, Tex) was used for band-limited imaging.
Blood Flow
Blood flow was measured using Laser Doppler flowmetry. For flank tumor experiments, 300-e diameter Laser Doppler probes (OxiFlo, Oxford Optronix, Oxford, UK) were simultaneously inserted into the tumor and quadriceps muscle of each animal. For window chamber experiments, Laser Doppler probes (LASERFLO, TSI, St. Paul, Minn) were positioned beneath the tumor window as previously described.22
SNO-Hb Assay
The determination of SNO-Hb concentrations is based on a reaction where cleavage of S-nitrosothiols yields a nitrosant that activates 4,5-diaminofluorescein (DAF-2, Sigma, St. Louis, Mo). Data acquired before drug infusion allowed us to determine basal SNO levels. The plasma half-life of SNO-Hb was calculated from exponential decay curves fitting each experimental curve. Hb concentration in plasma samples was determined by visible spectrophotometry. Expanded method is available online.
Vascular Reactivity In Vivo
Tumor-feeding arterioles in window chambers were visualized using transillumination. Arteriolar diameters were measured from videotaped images using an image-shearing monitor (IPM Inc., San Diego, Calif).
Mean Arterial Pressure and Heart Rate
Mean arterial pressure (MAP) and heart rate (HR) were measured with a blood pressure analyzer (Digi-Med, Micro-Med, Louisville, Ky) connected to the femoral artery cannula, as previously described.20
Statistical Analyses
All values are shown as means±S.E. Time curves were normalized to the baseline before infusion. ‘N’ refers to the number of animals per group and ‘n’ to individual measurements. Repeated-measure 2-way ANOVA or Student’s t tests were used as indicated.
Results
SNO-Hb and Oxy-Hb IV Reduce Muscle and Tumor Perfusion in Normoxic Rats
We first examined whether IV infusion of human cell-free oxy-Hb and SNO-oxy-Hb (SNO-Hb) influenced muscle perfusion in the quadriceps of rats breathing room air (normoxia). SNO-Hb decreased muscle perfusion more than Hb when compared with albumin (Figure 1A, P=0.055 and 0.3 versus albumin, respectively, 2-way ANOVA). We then determined the pharmacological effects of these molecules in flank tumors of normoxic rats. In contrast to albumin, Hb induced a rapid and sustained decrease in tumor blood flow (Figure 1B, P<0.01 versus albumin, 2-way ANOVA). Unexpectedly, i.v. infusion of SNO-Hb also reduced perfusion (Figure 1C, P<0.05 versus SNO-albumin, 2-way ANOVA). Reductions in flow caused by oxy-Hb and SNO-Hb were not different (P>0.05, 2-way ANOVA). However, the effects of the 2 Hbs were quite different from the small increase in tumor perfusion that we observed following albumin or SNO-albumin infusion i.v. (Figure 1B and 1C). The effects of SNO-albumin and albumin were not different (P>0.05, 2-way ANOVA).
Hyperoxia Prolongs the Half-life and Modulates the Pressor Activity of Cell-Free SNO-Hb
Based on previous observations,2 we reasoned that an increase in plasma oxygenation would stabilize SNO-Hb, increase peripheral delivery and thereby increase the bioactivity in tumors. We aimed to determine whether the plasma half-life of SNO-Hb i.v. was prolonged in rats breathing 100% O2 (hyperoxia) versus room air. Blood gas measurements showed that breathing 100% O2 induced significant increases in venous and arteriolar blood pO2, pCO2, and endogenous Hb oxygen saturation. Changes in Hb saturation were identical in femoral vein and tumor venules. Changes were less pronounced in tumor-feeding arterioles versus the femoral artery, as expected from blood deoxygenation along the arterial tree and at the tumor margin (see Table in the online supplement).
Because of low concentration (6.5 eol/L) and alteration in the tetramer/dimer ratio as a function of concentration and O2 saturation (dimer/tetramer equilibrium is 1000-fold lower in deoxy-Hb), it was not possible to directly measure the oxygen saturation of exogenous Hbs. Instead, we measured plasma SNO-Hb concentration versus time following infusion using a DAF-2 assay. The fitting of individual experimental data with exponential decay curves (R2=0.87±0.05 and 0.90±0.03 for normoxic and hyperoxic conditions, respectively) revealed a 2-fold increase in the half-life of the S-nitrosyl group in SNO-Hb in the plasma of hyperoxic versus normoxic rats (Figure 2, P<0.01, Student’s t test). We found no influence of O2 on the rate of SNO-Hb protein clearance (data not shown). Spectrophotometric measurements revealed that hyperoxia did not affect the rate of protein clearance during the period of the experiments (half-life 30 minutes, independent of O2 supply) (data not shown). Thus, our results, together with previous observations,23 suggest that the effects of O2 are mediated by an allosteric mechanism that promotes the R structure.
The effects of hyperoxia alone on tumor and muscle perfusion ranged from a transient decrease (that resumed within 5 minutes, ie, before drug infusion) to no change. As expected from these observations (and in agreement with prior publications),2,3 hyperoxia abolished the reduction in tumor perfusion that we observed after i.v. infusion of SNO-Hb in normoxic rats (Figure 3A, P<0.005, 2-way ANOVA). In contrast to SNO-Hb, and native Hb reduced tumor perfusion independently of O2 supply (Figure 3B, P>0.05, 2-way ANOVA).
Hyperoxia Unmasks the Systemic Pressor Activity of Cell-Free SNO-Hb
We then sought to identify the relative contribution of changes in vascular activity versus systemic hemodynamics in the tumor perfusion response to SNO-Hb. In window chambers, i.v. infusion of SNO-Hb induced no significant change in the diameter of feeding arterioles in normoxic or hyperoxic rats (Figure 4A, P>0.05, 2-way ANOVA). As reported in our previous studies,20,22 hyperoxia alone had no vascular effect (compare t5 to t=0 in Figure 4A, white boxes). Interestingly, in the same rats, we simultaneously documented that SNO-Hb induced a potent decrease in tumor perfusion in normoxic animals, and that this decrease was prevented in hyperoxic rats (data not shown). Thus, these results were identical to what we observed in flank tumors (Figure 3A).
Changes in the tumor blood flow were paralleled by changes in the perfusion of the leg quadriceps muscle: the decrease in muscle blood flow after delivery of SNO-Hb under normoxia was prevented under hyperoxia (Figure 4B, P<0.05, 2-way ANOVA). In contrast, hyperoxia did not modify the effect of oxy-Hb in muscles (data not shown).
Together, these observations suggest that SNO-Hb might indirectly modulate tumor perfusion in an O2-dependent manner as a consequence of changes in mean arterial pressure (MAP) and the heart rate (HR). We observed no change in MAP (Figure 4C) or HR (Figure 4D) when SNO-Hb was infused i.v. in normoxic animals (Student’s t test versus values at t=0). However, under hyperoxia, we measured a significant increase in the MAP as soon as 10 minutes after the infusion of SNO-Hb i.v. (Figure 4C, P<0.05 from t+10 to t+20, Student’s t test versus value at t=0). There was no significant effect on HR (Figure 4D, P>0.05 for all values versus value at t=0, Student’s t test). Interestingly, in the same conditions, oxy-Hb had no effect on MAP during normoxia or hyperoxia (Figure 4E, P>0.05 for all values versus value at t=0, Student’s t test). Also, in contrast to SNO-Hb, oxy-Hb induced bradycardia as soon as 5 minutes after infusion (Figure 4D, P<0.05 from t+5 to t+20, Student’s t test versus value at t=0), which was more severe in hyperoxic than in normoxic rats. Hyperoxia alone induced no significant changes in MAP (+1%) or HR (eC3%) (P>0.05, N=19, Student’s t test).
The Route of Administration Impacts the Pressor Activity of Cell-Free SNO-Hb
The exquisite O2 dependence of the bioactivity of SNO-Hb prompted us to check whether the physiological difference of blood oxygenation between veins (femoral vein infusion) and arteries (left carotid artery infusion) was sufficient to modulate the pressor activity of SNO-Hb. In normoxic animals, despite a trend toward preservation of the flow, the decrease in tumor perfusion after i.a. infusion of SNO-Hb was not significantly different compared with i.v. infusion (Figure 5A, P=0.09, 2-way ANOVA). However, in contrast to i.v. SNO-Hb, the pressor activity of i.a. SNO-Hb was not altered when this experiment was repeated in hyperoxic rats (P>0.05 versus SNO-Hb i.a. room air, 2-way ANOVA). The lack of O2 dependence contrasted with the strong O2-dependent changes in tumor perfusion for i.v. infusions (these curves are provided in Figure 5A for comparison).
Furthermore, O2 concentration in the breathing gas did not modify the perfusion of the quadriceps muscle when SNO-Hb was delivered i.a. (Figure 5B, P>0.05, 2-way ANOVA). As with the tumor results, the effects contrasted with the changes observed following i.v. injection (Figure 4B). Muscle blood flow remained unaltered after i.a. infusion of SNO-Hb, independently of O2 delivery (Figure 5B).
MAP was highly sensitive to SNO-Hb i.a.; we documented a significant increase in MAP as soon as 5 minutes and 10 minutes after drug delivery to normoxic and hyperoxic rats, respectively (Figure 5C, P<0.05 versus value at t=0, Student’s t test). The degree of hypertension, however, was unaffected by blood pO2. In the same set of experiments, HR remained unchanged (Figure 5D). In control experiments, i.a. and i.v. infusion of albumin (used to monitor volume effects on the baroreceptor) did not induce any significant or differential change in tumor perfusion, MAP, or HR of tumor-bearing rats (data not shown, all P>0.05 for albumin i.a. versus albumin i.v., 2-way ANOVA).
Discussion
Despite molecular modifications that successfully overcome the toxicity of unstable Hb tetramers,12 vasoconstriction remains a major limitation in the clinical use of Hb-based blood substitutes.13 Cell-free Hb preparations are devoid of the SNO that normally serves to counter their vasoconstrictive effects in vivo.23 We therefore investigated whether SNO reconstitution of Hb could reverse the vasoconstrictor activity of oxy-Hb, and whether SNO-Hb could be manipulated allosterically to maximize O2 delivery. In rats breathing room air (normoxia), SNO-Hb induced a greater decrease in tissue perfusion than native Hb (Figure 1A and Table 1). These results suggested that oxy-Hb and SNO-Hb operate by different mechanisms. Whereas the effects of oxy-Hb could be readily explained by NO scavenging (the "oxyhemoglobin," or met-Hb forming-reaction), those of SNO-Hb were consistent with other NO donors, which paradoxically also decrease tissue perfusion. NO-mediated dilation of healthy blood vessels upstream of tumors creates shunts that divert blood away from the tumors (vascular steal). By raising the concentration of the inhaled O2, SNO-Hb bioactivity can be targeted to more distally blood vessels, and elicits improvements in tumor blood flow.
We and recently others1,3,8 have proposed that SNO-Hb is formed in concert with O2 uptake in the lung, and that the release of vasodilator NO from SNO-Hb is favored at low pO2 (ie, in deoxygenated tissues and arterioles). The presence of diffusional barriers that attenuate endothelium-derived NO entry into the RBC represents an additional mechanism by which NO bioactivity may be preserved in large vessels.4 But whereas diffusional barriers are biologically important in limiting overall NO consumption, they play a little role in hypoxic vasodilation, which is subserved by erythrocytic SNO-Hb in microvessels3 (and is unaffected by NOS inhibitors and preserved in eNOSeC/eC animals).6 Thus, in the microcirculation (where RBC are in close contact with the vessel wall), the barrier cannot provide an explanation for precise control of NO bioactivity.
To further test the hypothesis of oxygen-dependent regulation of NO release, we investigated the activity of cell-free SNO-Hb versus native Hb in tumors. We used the well-characterized R3230Ac rat mammary tumor model20,24,25 as a means to observe low but biologically significant pO2 conditions. In normoxic rats, SNO-Hb i.v. lowered perfusion in tumors to the same extent as in muscles (Table 1). In tumors, oxy Hb exhibited a similar decrease in perfusion, whereas albumin (and to a similar degree SNO-albumin) tended to increase perfusion slightly (Figure 1B and 1C). Interestingly, the SNO-Hb-induced decrease in tumor perfusion was not associated with tumor-feeding vessel constriction or changes in MAP or HR (Table 1). It can therefore be attributed to either: 1) vasoconstriction (NO scavenging) of microvessels within the tumor, or 2) vasodilation (SNO release) of vessels in parallel tissues that creates a vascular steal. Because of ongoing angiogenesis, most vessels downstream of feeding arterioles in fast-growing rodent tumors lack structural elements for vasoactivity, and they lack functional endothelial NO synthase.26 Hence, during normoxia, i.v. SNO-Hb must decrease tumor perfusion through vascular steal. This is consistent with the previous observation that, although oxy-Hb provokes a decrease in healthy muscle perfusion (as expected from a NO scavenger), it increases perfusion in muscle surrounding the R3230AC tumor (a NO donor-like response) consistent with a steal effect.25
To gain further mechanistic insight, we reasoned that manipulation of blood oxygenation would impact SNO-Hb bioactivity. As reported in our previous studies,20,22 normobaric 100% oxygen breathing (hyperoxia) induced a significant increase in venous and arteriolar pO2. Although the delivery of pure oxygen sometimes reduced tumor and muscle perfusion (see Figure 3A, 3B, 5A, and 5B between t=eC5 and t=0), this effect was transient and returned to baseline within <5 minutes. Thus, the vasoactive properties of oxygen per se did not affect the response to SNO-Hb or oxy-Hb. On i.v. infusion in hyperoxic animals, SNO modified Hb successfully opposed the reduction in perfusion created by Hb in tumors (compare Figure 3A and 3B). Increases in MAP were seen in response to SNO-Hb during 100% O2 breathing, but there was no change in tumor/muscle perfusion or HR (Table 1). By comparison, the reduction in tumor perfusion by native Hb remained unaffected by hyperoxia (Figure 3B). This is likely accounted for by lowered systemic perfusion and bradycardia, not by changes in MAP (Table 1).
Hb dissociation from tetramers to dimers depends on Hb concentration and oxygen tension, and NO release from dimers is unresponsive to allosteric effectors (eg, O2).27 Assuming a plasma concentration of 6.5 eol/L, 35% of SNO-Hb would be tetramers (KD for R and T conformers are 3 eol/L and 3 nmol/L, respectively) at room air and the amount would increase precipitously as O2 tension declines - the majority would be tetramers at tissue pO2. Is there enough tetrameric SNO-Hb to sense oxygen and regulate NO delivery Evidently yes, as we documented a finely tuned regulation of SNO-Hb bioactivity by oxygen tension (NO release depends on local pO2) that contrasted with the O2-independent behavior of oxy-Hb. We further demonstrated that the plasma half-life of SNO in Hb was more than doubled during hyperoxia whereas protein clearance remained unaffected. Thus, the prolonged half-life of SNO at high pO2 is consistent with the expectation that release of NO bioactivity is disfavored in the R structure.1,2 Collectively these data provide a definitive demonstration of allosteric regulation by O2 of NO delivery from SNO-Hb.
During hyperoxia SNO-Hb is greatly stabilized, surviving arteriovenous transit. SNO-Hb bioactivity was thus indistinguishable following hyperoxic i.v. and normoxic i.a. infusions (Table 1). To address a point of confusion, we note that SNO-Hb will exert activity at both high and low pO2 but that this activity will be potentiated at low pO2.23 Effects will therefore manifest in deoxygenated vessels. Consistent with this interpretation, hyperoxia had no effect on i.a. infusions of SNO-Hb but markedly altered i.v. responses.
On close inspection, our data unravel a central effect of SNO-Hb on control of hemodynamics. An increase in MAP was observed following i.a. infusion during normoxia, and both i.v. and i.a. infusions during hyperoxia. Native Hb (hyperoxia, i.v.) did not produce increases in MAP (Table 1) under any condition. Baroreceptor activity opposes increases in MAP and resistance to flow. Native Hb activated the baroreceptor, which opposed changes in MAP by decreasing the HR. In contrast, strikingly, SNO-Hb raised the MAP without altering perfusion (in the organs that we monitored) or HR. The systemic pressor activity of SNO-Hb results from a direct inhibition of the baroreceptor reflex. Indeed, at higher concentrations (1 eol/kg, i.v., normoxia), it has been reported that both SNO-Hb and oxy-Hb overcome baroreceptor buffering capacities (presumably indirectly through NO scavenging and peripheral vasoconstriction).2 Thus, SNO-Hb activity in hyperbaric hyperoxia is sufficient to inhibit the baroreceptor. As a neurotransmitter, NO is known to regulate baroreceptor activity through sympathetic afferent nerve fibers.28 Exogenous NO has an inhibitory action on the baroreceptor, albeit only at high concentrations, and under such circumstances that the baroreceptor activity recovers within seconds.29 Unlike classical NO activity, SNO-Hb bioactivity is mediated by species distinct from N·O itself (eg, low-mass SNOs).27,30,31 It is noteworthy that other endogenous S-nitrosothiols (eg, SNO-cysteine) can suppress baroreceptor activity independently of cGMP generation (the mediator of classical nitrosovasodilator activity).32 Stereoselective recognition sites in the baroreceptor vasculature could mediate baroreceptor inhibition by S-nitrosylated species such as SNO-cysteine and SNO-Hb.33 Expanded discussion is available online.
In conclusion, we have shown that SNO reconstitution of Hb successfully overcomes the reduction in tumor perfusion created by Hb itself. This activity involves the allosteric control of NO release by O2. The hemodynamic effects of SNO-Hb are a composite of central activity (baroreceptor inhibition) and allostery-facilitated NO release in the peripheral circulation. In contrast, native Hb, at concentrations characteristic of hemolytic states, has no central pressor effect. These results are summarized in Table 2.
Whereas the concentration of Hb we used is low from the O2 delivery standpoint, it greatly exceeds the concentration of any endogenous NO. In light of our results, development of safe Hb-based blood substitutes might include approaches that not only preserve Hb allostery (eg, crosslinking to prevent dissociation into dimmers) but also reconstitute SNO content. Reoxygenation of hypoxic tissues may benefit from both the central pressor and blood flow increasing effects elicited by SNO-Hb. Further, the treatment of trauma or septic patients might involve manipulation of SNO-Hb allostery to limit the excessive release of NO.
Acknowledgments
This work was supported by grants from the NIH/NCI CA40355 (M.W.D.), the Howard Hughes Medical Institute (J.S.S.), the NIH Division of Research Resources COBRE University of Puerto Rico Protein Research Center (Grant 5P20RR016439) (J.B.), and a collaborative travel award from the North Atlantic Treaty Organization (NATO) (G.T.). P.S. is a fellow of the Belgian American Educational Foundation (BAEF) and a ‘Fonds National de la Recherche Scientifique’ (FNRS) post-doctoral fellow detached from the University of Louvain Medical School (UCL, Unit of Pharmacology and Therapeutics, Brussels, Belgium). T.M. received support from the Institute for Medical Research, Inc, Durham Veteran’s Affairs Medical Center. The authors appreciate the assistance of Christine Baudelet and Pelzer O. Doar.
J.S.S. has a financial interest in Nitrox LLC, and is a paid consultant.
Both authors contributed equally to this work.
References
Jia L, Bonaventura C, Bonaventura J, Stamler JS. S-nitrosohaemoglobin: a dynamic activity of blood involved in vascular control. Nature. 1996; 380: 221eC226.
Stamler JS, Jia L, Eu JP, McMahon TJ, Demchenko IT, Bonaventura J, Gernert K, Piantadosi CA. Blood flow regulation by S-nitrosohemoglobin in the physiological oxygen gradient. Science. 1997; 276: 2034eC2037.
McMahon TJ, Moon RE, Luschinger BP, Carraway MS, Stone AE, Stolp BW, Gow AJ, Pawloski JR, Watke P, Singel DJ, Piantadosi CA, Stamler JS. Nitric oxide in the human respiratory cycle. Nat Med. 2002; 8: 711eC717.
Crawford JH, White CR, Patel RP. Vasoactivity of S-nitrosohemoglobin: role of oxygen, heme, and NO oxidation states. Blood. 2003; 101: 4408eC4415.
Gladwin MT, Lancaster JR Jr, Freeman BA, Schechter AN. Nitric oxide’s reactions with hemoglobin: a view through the SNO-storm. Nat Med. 2003; 9: 496eC500.
Gonzalez-Alonso J, Olsen DB, Saltin B. Erythrocyte and the regulation of human skeletal muscle blood flow and oxygen delivery: role of circulating ATP. Circ Res. 2002; 91: 1046eC1055.
Cosby K, Partovi KS, Crawford JH, Patel RP, Reiter CD, Martyr S, Yang BK, Waclawiw MA, Zalos G, Xu X, Huang KT, Shields H, Kim-Shapiro DB, Schechter AN, Cannon RO3rd, Gladwin MT. Nitrite reduction to nitric oxide by deoxyhemoglobin vasodilates the human circulation. Nat Med. 2003; 9: 1498eC1505.
James PE, Lang D, Tufnell-Barret T, Milsom AB, Frenneaux MP. Vasorelaxation by red blood cells and impairment in diabetes: reduced nitric oxide and oxygen delivery by glycated hemoglobin. Circ Res. 2004; 94: 976eC983.
Datta B, Tufnell-Barrett T, Bleasdale RA, Jones CJ, Beeton I, Paul V, Frenneaux M, James P. Red blood cell nitric oxide as an endocrine vasoregulator: a potential role in congestive heart failure. Circulation. 2004; 109: 1339eC1342.
Savitsky JP, Doczi J, Black J, Arnold JD. A clinical safety trial of stroma-free hemoglobin. Clin Pharmacol Ther. 1978; 23: 73eC80.
Bunn HF, Esham WT, Bull RW. The renal handling of hemoglobin. I. Glomerular filtration. J Exp Med. 1969; 129: 909eC923.
Alayash AI. Oxygen therapeutics: can we tame haemoglobin Nat Rev Drug Discov. 2004; 3: 152eC159.
Deem S, Kim JU, Manjula BN, Acharya AS, Kerr ME, Patel RP, Gladwin MT, Swenson ER. Effects of S-nitrosation and cross-linking of hemoglobin on hypoxic pulmonary vasoconstriction in isolated rat lungs. Circ Res. 2002; 91: 626eC632.
Doyle MP, Hoekstra JW. Oxidation of nitrogen oxides by bound dioxygen in hemoproteins. J Inorg Biochem. 1981; 14: 351eC358.
Eich RF, Li T, Lemon DD, Doherty DH, Curry SR, Aitken JF, Mathews AJ, Johnson KA, Smith RD, Phillips GNJR, Olson JS. Mechanism of NO-induced oxidation of myoglobin and hemoglobin. Biochemistry. 1996; 35: 6976eC6983.
Jourd’heuil D, Gray L, Grisham MB. S-nitrosothiol formation in blood of lipopolysaccharide-treated rats. Biochem Biophys Res Commun. 2000; 273: 22eC26.
Nakai K, Sakuma I, Togashi H, Yoshioka M, Sugawara T, Satoh H, Kitabatake A. S-nitrosylated polyethylene glycol-conjugated hemoglobin derivative as a candidate material for oxygen therapeutics. Adv Exp Med Biol. 2003; 519: 207eC216.
Hilf R, Michel I, Bell C, Freeman JJ, Borman A. Biochemical and morphologic properties of a new lactating mammary tumor line in the rat. Cancer Res. 1965; 25: 286eC299.
McMahon TJ, Stamler JS. Concerted nitric oxide/oxygen delivery by hemoglobin. Methods Enzymol. 1999; 301: 99eC114.
Dewhirst MW, Ong ET, Rosner GL, Rehmus SW, Shan S, Braun RD, Brizel DM, Secomb TW. Arteriolar oxygenation in tumour and subcutaneous arterioles: effects of inspired air oxygen content. Br J Cancer Suppl. 1996; 27: S241eCS246.
Sorg BS, Moeller BJ, Donovan O, Cao Y, Dewhirst MW. Hyperspectral imaging of hemoglobin saturation in tumor microvasculature and tumor hypoxia development. J Biomed Opt. 2005 (In press).
Dunn TJ, Braun RD, Rhemus WE, Rosner GL, Secomb TW, Tozer GM, Chaplin DJ, Dewhirst MW. The effects of hyperoxic and hypercarbic gases on tumour blood flow. Br J Cancer. 1999; 80: 117eC126.
Singel DJ, Stamler JS. Chemical physiology of blood flow regulation by red blood cells: the role of nitric oxide and S-nitrosohemoglobin. Annu Rev Physiol. 2005; 67: 99eC145.
Shan SQ, Rosner GL, Braun RD, Hahn J, Pearce C, Dewhirst MW. Effects of diethylamine/nitric oxide on blood perfusion and oxygenation in the R3230Ac mammary carcinoma. Br J Cancer. 1997; 76: 429eC437.
Hahn JS, Braun RD, Dewhirst MW, Shan S, Snyder SA, Taube JM, Ong ET, Rosner GL, Dodge RK, Bonaventura J, Bonaventura C, DeAngelo J, Meyer RE. Stroma-free human hemoglobin A decreases R3230Ac rat mammary adenocarcinoma blood flow and oxygen partial pressure. Radiat Res. 1997; 147: 185eC194.
Sonveaux P, Dessy C, Brouet A, Jordan BF, Gregoire V, Gallez B, Balligand JL, Feron O. Modulation of the tumor vasculature functionality by ionizing radiation accounts for tumor radiosensitization and promotes gene delivery. FASEB J. 2002; 16: 1979eC1981.
Lipton AJ, Johnson MA, Macdonald T, Lieberman MW, Gozal D, Gaston B. S-nitrosothiols signal the ventilatory response to hypoxia. Nature. 2001; 413: 171eC174.
Paton JF, Deuchars J, Ahmad Z, Wong LF, Murphy D, Kasparov S. Adenoviral vector demonstrates that angiotensin II-induced depression of the cardiac baroreflex is mediated by endothelial nitric oxide synthase in the nucleus tractus solitarii of the rat. J Physiol. 2001; 531 (Pt 2): 445eC58.
Harada S, Tokunaga S, Momohara M, Masaki H, Tagawa T, Imaizumi T, Takeshita A. Inhibition of nitric oxide formation in the nucleus tractus solitarius increases renal sympathetic nerve activity in rabbits. Circ Res. 1993; 72: 511eC516.
Pawloski JR, Swaminathan RV, Stamler JS. Cell-free and erythrocytic S-nitrosohemoglobin inhibits human platelet aggregation. Circulation. 1998; 97 (3): 263eC267.
Liu L, Yan Y, Zeng M, Zhang J, Hanes MA, Ahearn G, McMahon TJ, Dickfeld T, Marshall HE, Que LG, Stamler JS. Essential roles of S-nitrosothiols in vascular homeostasis and endotoxic shock. Cell. 2004; 116: 617eC628.
Matsuda T, Bates JN, Lewis SJ, Abboud FM, Chapleau MW. Modulation of baroreceptor activity by nitric oxide and S-nitrosocysteine. Circ Res. 1995; 76: 426eC433.
Davisson RL, Travis MD, Bates JN, Lewis SJ. Hemodynamic effects of L- and D-S-nitrosocysteine in the rat. Stereoselective S-nitrosothiol recognition sites. Circ Res. 1996; 79: 256eC262.(Pierre Sonveaux, Andrew M)
Nicholas School of the Environment and Earth Sciences, Duke University Marine Laboratory
University of Puerto Rico NIH COBREII Protein Research Center, Beaufort, NC and Mayaguez PR00681 (J.B.)
Tumor Microcirculation Group, Gray Cancer Institute, Mount Vernon Hospital, Northwood, Middlesex, UK (G.M.T).
Abstract
In erythrocytes, S-nitrosohemoglobin (SNO-Hb) arises from S-nitrosylation of oxygenated hemoglobin (Hb). It has been shown that SNO-Hb behaves as a nitric oxide (NO) donor at low oxygen tensions. This property, in combination with oxygen transport capacity, suggests that SNO-Hb may have unique potential to reoxygenate hypoxic tissues. The present study was designed to test the idea that the allosteric properties of SNO-Hb could be manipulated to enhance oxygen delivery in a hypoxic tumor. Using Laser Doppler flowmetry, we showed that SNO-Hb infusion to animals breathing 21% O2 reduced tumor perfusion without affecting blood pressure and heart rate. Raising the pO2 (100% O2) slowed the release of NO bioactivity from SNO-Hb (ie, prolonged the plasma half-life of the SNO in Hb), preserved tumor perfusion, and raised the blood pressure. In contrast, native Hb reduced both tumor perfusion and heart rate independently of the oxygen concentration of the inhaled gas, and did not elicit hypertensive effects. Window chamber (to image tumor arteriolar reactivity in vivo) and hemodynamic measurements indicated that the preservation of tissue perfusion by micromolar concentrations of SNO-Hb is a composite effect created by reduced peripheral vascular resistance and direct inhibition of the baroreceptor reflex, leading to increased blood pressure. Overall, these results indicate that the properties of SNO-Hb are attributable to allosteric control of NO release by oxygen in central as well as peripheral issues.
Key Words: nitric oxide hemoglobin oxygen hemodynamics blood flow
Introduction
Hemoglobin (Hb) of red blood cells (RBC) is a tetrameric protein composed of 2 and 2 chains, each containing a heme prosthetic group. One and 1 chain is combined in stable dimers, and 2 dimers are more loosely associated to form tetramers. The physiological role of Hb depends on its ability to reversibly bind O2 at its heme iron centers. This transport capacity is governed by a cycle of allosteric transitions in which Hb assumes the R (relaxed, high O2 affinity) conformation to bind O2 in the lungs and, on partial deoxygenation, the T (tense, low O2 affinity) conformation to efficiently deliver O2 to peripheral tissues. This transition also controls the reactivity of 2 conserved cysteines on the chains (Cys93). Thiols of the cysteines can react with the potent vasodilator nitric oxide (NO) to form S-nitrosohemoglobin(SNO-Hb) in the R conformation, and preferentially unload SNO in the T conformation.1eC3 It has therefore been proposed that Hb would carry NO equivalents from the lungs to the periphery, thereby bringing tissue blood flow in line with oxygen demand. Although mechanisms are debated,4eC7 some consensus has been reached, namely that Hb can act as an oxygen sensor and oxygen-dependent transducer of NO bioactivity,7eC9 and the direct coupling between SNO and O2 content of Hb has been recently affirmed in intact human erythrocytes (Doctor et al, unpublished observations).
Outside the red blood cell (RBC), Hb has toxic effects that prevent its use as a blood substitute in humans.10 In plasma, Hb rapidly dissociates to dimers, leading to hemoglobinuria (renal injury) and extravasation (osmotic disorders).11 Several types of conjugated and intramolecularly cross-linked Hbs have been developed that reduce Hb dissociation, thereby lowering the risks for such toxicity.12 These modified Hbs also show prolonged vascular retention times, making them promising blood substitute candidates. However, clinical safety is still not achieved, mainly because modified Hbs retain the vasoconstrictive activity of cell-free Hb.13 Ferrous Hb and oxy-Hb can react with NO to produce iron-nitrosyl Hb, and oxidized Hb and nitrate ions (met-Hb reaction), respectively.14,15 The Hb concentrations required to exert such activity have not been carefully examined, and the extent to which vasodilation is involved (under more physiological circumstances, Hb can also react with NO or S-nitrosothiols to yield SNO-Hb)16 remains a matter of debate. Lower doses of Hb-based blood substitutes do not typically raise blood pressure.
To date, SNO-Hb, which combines the capacity to carry oxygen and the potential to act as a hypoxia-activated NO donor, has not yet been fully tested as a RBC substitute.17 In the present work, we hypothesized that S-nitrosylation of cell-free Hb would reduce the vasoconstrictive effects of tetrameric Hb. We also hypothesized that blood oxygenation, through allostery-regulated control of NO release, would be a key determinant of the biological responses to SNO-Hb. These hypotheses were investigated in vivo using a physiological model of vascular hypoxia (perfused tumor vessels), in which pO2 was modulated in situ by varying inspired O2. This prospective study offers a new perspective on the development of transfusion substitutes, and establishes that delivery of NO bioactivity by SNO-Hb is regulated by O2 saturation.
Materials and Methods
Rats and Tumors
Seventy-four female Fischer 344 rats (Charles River, Raleigh, NC), bearing or not bearing the syngeneic mammary adenocarcinoma R3230AC,18 were used in all experiments. Tumor pieces from donor rats were implanted in anesthetized animals (100 mg/kg ketamine and 10 mg/kg xylazine i.p.). Tumors were grown in the subcutis of the left lateral quadriceps muscle or, for window chamber experiments, between 2 fascial layers on the back of rats. Animals were then randomly assigned to a treatment group. Further interventions/observations were performed on anesthetized animals (50 mg/kg pentobarbital i.p.) maintained at 37°C on a temperature-controlled thermal blanket. Where indicated, some rats were fitted with a facemask for breathing 100% O2. All experiments were approved by Duke IACUC.
Drugs
SNO-Hb was synthesized through reaction of purified human hemoglobin A0 (Apex Bioscience, Durham, NC) with SNO-cysteine in nonacidic conditions, as previously reported19 (expanded method at http://www.circresaha.org). SNO-albumin was synthesized by the reaction of an equal volume of bovine serum albumin (0.2 mmol/L in HCl 0.1 mmol/L, 0.5 mmol/L EDTA) with 0.2 mmol/L NaNO2 for 30 minutes. All solutions (diluted at 200 eol/L in PBS pH 7.4, 0.5 mmol/L EDTA) were kept on ice or frozen and protected from light until administration. Less than 2-weeks old solutions were infused at a dose of 200 nmol/kg in 0.25 mL, immediately followed by a 0.25 mL delivery of saline. Assuming a plasma volume of 3 mL per 100 g in rats and no significant partition in blood cells, this dose would achieve a maximum plasma concentration of 6.5 eol/L for all drugs. To avoid volume-induced alterations of hemodynamics, solutions were infused at 0.5 mL/min from time 0.
Surgery
For i.v. infusions, the femoral artery and vein were cannulated and drugs were infused through the venous cannula. For i.a. infusions, the femoral and left carotid arteries were cannulated and drugs were infused through the carotid cannula. For window chamber experiments, the 2 window frames were surgically positioned in the dorsal skin flap, followed by tumor cell transplantation, as previously described.20 Following several days of tumor growth, direct visualization of tumor-feeding arterioles was performed using intravital microscopy.
Blood Gas and Hemoglobin Saturation Measurements in Tumor Microvessels
Blood gases and Hb oxygen saturation were measured using a 1640 oximeter coupled to a 482 cooximeter (Instrumentation Laboratories, Lexington, Mass) from 0.5 mL samples of femoral arterial and venous blood. For tumor microvessels in window chambers, Hb saturation was calculated from transmission optical measurements of vascular absorbance using a hyperspectral imaging technique.21 A liquid crystal tunable filter (CRI, Inc., Woburn, Mass) placed in front of a CCD camera (DVC Company, Austin, Tex) was used for band-limited imaging.
Blood Flow
Blood flow was measured using Laser Doppler flowmetry. For flank tumor experiments, 300-e diameter Laser Doppler probes (OxiFlo, Oxford Optronix, Oxford, UK) were simultaneously inserted into the tumor and quadriceps muscle of each animal. For window chamber experiments, Laser Doppler probes (LASERFLO, TSI, St. Paul, Minn) were positioned beneath the tumor window as previously described.22
SNO-Hb Assay
The determination of SNO-Hb concentrations is based on a reaction where cleavage of S-nitrosothiols yields a nitrosant that activates 4,5-diaminofluorescein (DAF-2, Sigma, St. Louis, Mo). Data acquired before drug infusion allowed us to determine basal SNO levels. The plasma half-life of SNO-Hb was calculated from exponential decay curves fitting each experimental curve. Hb concentration in plasma samples was determined by visible spectrophotometry. Expanded method is available online.
Vascular Reactivity In Vivo
Tumor-feeding arterioles in window chambers were visualized using transillumination. Arteriolar diameters were measured from videotaped images using an image-shearing monitor (IPM Inc., San Diego, Calif).
Mean Arterial Pressure and Heart Rate
Mean arterial pressure (MAP) and heart rate (HR) were measured with a blood pressure analyzer (Digi-Med, Micro-Med, Louisville, Ky) connected to the femoral artery cannula, as previously described.20
Statistical Analyses
All values are shown as means±S.E. Time curves were normalized to the baseline before infusion. ‘N’ refers to the number of animals per group and ‘n’ to individual measurements. Repeated-measure 2-way ANOVA or Student’s t tests were used as indicated.
Results
SNO-Hb and Oxy-Hb IV Reduce Muscle and Tumor Perfusion in Normoxic Rats
We first examined whether IV infusion of human cell-free oxy-Hb and SNO-oxy-Hb (SNO-Hb) influenced muscle perfusion in the quadriceps of rats breathing room air (normoxia). SNO-Hb decreased muscle perfusion more than Hb when compared with albumin (Figure 1A, P=0.055 and 0.3 versus albumin, respectively, 2-way ANOVA). We then determined the pharmacological effects of these molecules in flank tumors of normoxic rats. In contrast to albumin, Hb induced a rapid and sustained decrease in tumor blood flow (Figure 1B, P<0.01 versus albumin, 2-way ANOVA). Unexpectedly, i.v. infusion of SNO-Hb also reduced perfusion (Figure 1C, P<0.05 versus SNO-albumin, 2-way ANOVA). Reductions in flow caused by oxy-Hb and SNO-Hb were not different (P>0.05, 2-way ANOVA). However, the effects of the 2 Hbs were quite different from the small increase in tumor perfusion that we observed following albumin or SNO-albumin infusion i.v. (Figure 1B and 1C). The effects of SNO-albumin and albumin were not different (P>0.05, 2-way ANOVA).
Hyperoxia Prolongs the Half-life and Modulates the Pressor Activity of Cell-Free SNO-Hb
Based on previous observations,2 we reasoned that an increase in plasma oxygenation would stabilize SNO-Hb, increase peripheral delivery and thereby increase the bioactivity in tumors. We aimed to determine whether the plasma half-life of SNO-Hb i.v. was prolonged in rats breathing 100% O2 (hyperoxia) versus room air. Blood gas measurements showed that breathing 100% O2 induced significant increases in venous and arteriolar blood pO2, pCO2, and endogenous Hb oxygen saturation. Changes in Hb saturation were identical in femoral vein and tumor venules. Changes were less pronounced in tumor-feeding arterioles versus the femoral artery, as expected from blood deoxygenation along the arterial tree and at the tumor margin (see Table in the online supplement).
Because of low concentration (6.5 eol/L) and alteration in the tetramer/dimer ratio as a function of concentration and O2 saturation (dimer/tetramer equilibrium is 1000-fold lower in deoxy-Hb), it was not possible to directly measure the oxygen saturation of exogenous Hbs. Instead, we measured plasma SNO-Hb concentration versus time following infusion using a DAF-2 assay. The fitting of individual experimental data with exponential decay curves (R2=0.87±0.05 and 0.90±0.03 for normoxic and hyperoxic conditions, respectively) revealed a 2-fold increase in the half-life of the S-nitrosyl group in SNO-Hb in the plasma of hyperoxic versus normoxic rats (Figure 2, P<0.01, Student’s t test). We found no influence of O2 on the rate of SNO-Hb protein clearance (data not shown). Spectrophotometric measurements revealed that hyperoxia did not affect the rate of protein clearance during the period of the experiments (half-life 30 minutes, independent of O2 supply) (data not shown). Thus, our results, together with previous observations,23 suggest that the effects of O2 are mediated by an allosteric mechanism that promotes the R structure.
The effects of hyperoxia alone on tumor and muscle perfusion ranged from a transient decrease (that resumed within 5 minutes, ie, before drug infusion) to no change. As expected from these observations (and in agreement with prior publications),2,3 hyperoxia abolished the reduction in tumor perfusion that we observed after i.v. infusion of SNO-Hb in normoxic rats (Figure 3A, P<0.005, 2-way ANOVA). In contrast to SNO-Hb, and native Hb reduced tumor perfusion independently of O2 supply (Figure 3B, P>0.05, 2-way ANOVA).
Hyperoxia Unmasks the Systemic Pressor Activity of Cell-Free SNO-Hb
We then sought to identify the relative contribution of changes in vascular activity versus systemic hemodynamics in the tumor perfusion response to SNO-Hb. In window chambers, i.v. infusion of SNO-Hb induced no significant change in the diameter of feeding arterioles in normoxic or hyperoxic rats (Figure 4A, P>0.05, 2-way ANOVA). As reported in our previous studies,20,22 hyperoxia alone had no vascular effect (compare t5 to t=0 in Figure 4A, white boxes). Interestingly, in the same rats, we simultaneously documented that SNO-Hb induced a potent decrease in tumor perfusion in normoxic animals, and that this decrease was prevented in hyperoxic rats (data not shown). Thus, these results were identical to what we observed in flank tumors (Figure 3A).
Changes in the tumor blood flow were paralleled by changes in the perfusion of the leg quadriceps muscle: the decrease in muscle blood flow after delivery of SNO-Hb under normoxia was prevented under hyperoxia (Figure 4B, P<0.05, 2-way ANOVA). In contrast, hyperoxia did not modify the effect of oxy-Hb in muscles (data not shown).
Together, these observations suggest that SNO-Hb might indirectly modulate tumor perfusion in an O2-dependent manner as a consequence of changes in mean arterial pressure (MAP) and the heart rate (HR). We observed no change in MAP (Figure 4C) or HR (Figure 4D) when SNO-Hb was infused i.v. in normoxic animals (Student’s t test versus values at t=0). However, under hyperoxia, we measured a significant increase in the MAP as soon as 10 minutes after the infusion of SNO-Hb i.v. (Figure 4C, P<0.05 from t+10 to t+20, Student’s t test versus value at t=0). There was no significant effect on HR (Figure 4D, P>0.05 for all values versus value at t=0, Student’s t test). Interestingly, in the same conditions, oxy-Hb had no effect on MAP during normoxia or hyperoxia (Figure 4E, P>0.05 for all values versus value at t=0, Student’s t test). Also, in contrast to SNO-Hb, oxy-Hb induced bradycardia as soon as 5 minutes after infusion (Figure 4D, P<0.05 from t+5 to t+20, Student’s t test versus value at t=0), which was more severe in hyperoxic than in normoxic rats. Hyperoxia alone induced no significant changes in MAP (+1%) or HR (eC3%) (P>0.05, N=19, Student’s t test).
The Route of Administration Impacts the Pressor Activity of Cell-Free SNO-Hb
The exquisite O2 dependence of the bioactivity of SNO-Hb prompted us to check whether the physiological difference of blood oxygenation between veins (femoral vein infusion) and arteries (left carotid artery infusion) was sufficient to modulate the pressor activity of SNO-Hb. In normoxic animals, despite a trend toward preservation of the flow, the decrease in tumor perfusion after i.a. infusion of SNO-Hb was not significantly different compared with i.v. infusion (Figure 5A, P=0.09, 2-way ANOVA). However, in contrast to i.v. SNO-Hb, the pressor activity of i.a. SNO-Hb was not altered when this experiment was repeated in hyperoxic rats (P>0.05 versus SNO-Hb i.a. room air, 2-way ANOVA). The lack of O2 dependence contrasted with the strong O2-dependent changes in tumor perfusion for i.v. infusions (these curves are provided in Figure 5A for comparison).
Furthermore, O2 concentration in the breathing gas did not modify the perfusion of the quadriceps muscle when SNO-Hb was delivered i.a. (Figure 5B, P>0.05, 2-way ANOVA). As with the tumor results, the effects contrasted with the changes observed following i.v. injection (Figure 4B). Muscle blood flow remained unaltered after i.a. infusion of SNO-Hb, independently of O2 delivery (Figure 5B).
MAP was highly sensitive to SNO-Hb i.a.; we documented a significant increase in MAP as soon as 5 minutes and 10 minutes after drug delivery to normoxic and hyperoxic rats, respectively (Figure 5C, P<0.05 versus value at t=0, Student’s t test). The degree of hypertension, however, was unaffected by blood pO2. In the same set of experiments, HR remained unchanged (Figure 5D). In control experiments, i.a. and i.v. infusion of albumin (used to monitor volume effects on the baroreceptor) did not induce any significant or differential change in tumor perfusion, MAP, or HR of tumor-bearing rats (data not shown, all P>0.05 for albumin i.a. versus albumin i.v., 2-way ANOVA).
Discussion
Despite molecular modifications that successfully overcome the toxicity of unstable Hb tetramers,12 vasoconstriction remains a major limitation in the clinical use of Hb-based blood substitutes.13 Cell-free Hb preparations are devoid of the SNO that normally serves to counter their vasoconstrictive effects in vivo.23 We therefore investigated whether SNO reconstitution of Hb could reverse the vasoconstrictor activity of oxy-Hb, and whether SNO-Hb could be manipulated allosterically to maximize O2 delivery. In rats breathing room air (normoxia), SNO-Hb induced a greater decrease in tissue perfusion than native Hb (Figure 1A and Table 1). These results suggested that oxy-Hb and SNO-Hb operate by different mechanisms. Whereas the effects of oxy-Hb could be readily explained by NO scavenging (the "oxyhemoglobin," or met-Hb forming-reaction), those of SNO-Hb were consistent with other NO donors, which paradoxically also decrease tissue perfusion. NO-mediated dilation of healthy blood vessels upstream of tumors creates shunts that divert blood away from the tumors (vascular steal). By raising the concentration of the inhaled O2, SNO-Hb bioactivity can be targeted to more distally blood vessels, and elicits improvements in tumor blood flow.
We and recently others1,3,8 have proposed that SNO-Hb is formed in concert with O2 uptake in the lung, and that the release of vasodilator NO from SNO-Hb is favored at low pO2 (ie, in deoxygenated tissues and arterioles). The presence of diffusional barriers that attenuate endothelium-derived NO entry into the RBC represents an additional mechanism by which NO bioactivity may be preserved in large vessels.4 But whereas diffusional barriers are biologically important in limiting overall NO consumption, they play a little role in hypoxic vasodilation, which is subserved by erythrocytic SNO-Hb in microvessels3 (and is unaffected by NOS inhibitors and preserved in eNOSeC/eC animals).6 Thus, in the microcirculation (where RBC are in close contact with the vessel wall), the barrier cannot provide an explanation for precise control of NO bioactivity.
To further test the hypothesis of oxygen-dependent regulation of NO release, we investigated the activity of cell-free SNO-Hb versus native Hb in tumors. We used the well-characterized R3230Ac rat mammary tumor model20,24,25 as a means to observe low but biologically significant pO2 conditions. In normoxic rats, SNO-Hb i.v. lowered perfusion in tumors to the same extent as in muscles (Table 1). In tumors, oxy Hb exhibited a similar decrease in perfusion, whereas albumin (and to a similar degree SNO-albumin) tended to increase perfusion slightly (Figure 1B and 1C). Interestingly, the SNO-Hb-induced decrease in tumor perfusion was not associated with tumor-feeding vessel constriction or changes in MAP or HR (Table 1). It can therefore be attributed to either: 1) vasoconstriction (NO scavenging) of microvessels within the tumor, or 2) vasodilation (SNO release) of vessels in parallel tissues that creates a vascular steal. Because of ongoing angiogenesis, most vessels downstream of feeding arterioles in fast-growing rodent tumors lack structural elements for vasoactivity, and they lack functional endothelial NO synthase.26 Hence, during normoxia, i.v. SNO-Hb must decrease tumor perfusion through vascular steal. This is consistent with the previous observation that, although oxy-Hb provokes a decrease in healthy muscle perfusion (as expected from a NO scavenger), it increases perfusion in muscle surrounding the R3230AC tumor (a NO donor-like response) consistent with a steal effect.25
To gain further mechanistic insight, we reasoned that manipulation of blood oxygenation would impact SNO-Hb bioactivity. As reported in our previous studies,20,22 normobaric 100% oxygen breathing (hyperoxia) induced a significant increase in venous and arteriolar pO2. Although the delivery of pure oxygen sometimes reduced tumor and muscle perfusion (see Figure 3A, 3B, 5A, and 5B between t=eC5 and t=0), this effect was transient and returned to baseline within <5 minutes. Thus, the vasoactive properties of oxygen per se did not affect the response to SNO-Hb or oxy-Hb. On i.v. infusion in hyperoxic animals, SNO modified Hb successfully opposed the reduction in perfusion created by Hb in tumors (compare Figure 3A and 3B). Increases in MAP were seen in response to SNO-Hb during 100% O2 breathing, but there was no change in tumor/muscle perfusion or HR (Table 1). By comparison, the reduction in tumor perfusion by native Hb remained unaffected by hyperoxia (Figure 3B). This is likely accounted for by lowered systemic perfusion and bradycardia, not by changes in MAP (Table 1).
Hb dissociation from tetramers to dimers depends on Hb concentration and oxygen tension, and NO release from dimers is unresponsive to allosteric effectors (eg, O2).27 Assuming a plasma concentration of 6.5 eol/L, 35% of SNO-Hb would be tetramers (KD for R and T conformers are 3 eol/L and 3 nmol/L, respectively) at room air and the amount would increase precipitously as O2 tension declines - the majority would be tetramers at tissue pO2. Is there enough tetrameric SNO-Hb to sense oxygen and regulate NO delivery Evidently yes, as we documented a finely tuned regulation of SNO-Hb bioactivity by oxygen tension (NO release depends on local pO2) that contrasted with the O2-independent behavior of oxy-Hb. We further demonstrated that the plasma half-life of SNO in Hb was more than doubled during hyperoxia whereas protein clearance remained unaffected. Thus, the prolonged half-life of SNO at high pO2 is consistent with the expectation that release of NO bioactivity is disfavored in the R structure.1,2 Collectively these data provide a definitive demonstration of allosteric regulation by O2 of NO delivery from SNO-Hb.
During hyperoxia SNO-Hb is greatly stabilized, surviving arteriovenous transit. SNO-Hb bioactivity was thus indistinguishable following hyperoxic i.v. and normoxic i.a. infusions (Table 1). To address a point of confusion, we note that SNO-Hb will exert activity at both high and low pO2 but that this activity will be potentiated at low pO2.23 Effects will therefore manifest in deoxygenated vessels. Consistent with this interpretation, hyperoxia had no effect on i.a. infusions of SNO-Hb but markedly altered i.v. responses.
On close inspection, our data unravel a central effect of SNO-Hb on control of hemodynamics. An increase in MAP was observed following i.a. infusion during normoxia, and both i.v. and i.a. infusions during hyperoxia. Native Hb (hyperoxia, i.v.) did not produce increases in MAP (Table 1) under any condition. Baroreceptor activity opposes increases in MAP and resistance to flow. Native Hb activated the baroreceptor, which opposed changes in MAP by decreasing the HR. In contrast, strikingly, SNO-Hb raised the MAP without altering perfusion (in the organs that we monitored) or HR. The systemic pressor activity of SNO-Hb results from a direct inhibition of the baroreceptor reflex. Indeed, at higher concentrations (1 eol/kg, i.v., normoxia), it has been reported that both SNO-Hb and oxy-Hb overcome baroreceptor buffering capacities (presumably indirectly through NO scavenging and peripheral vasoconstriction).2 Thus, SNO-Hb activity in hyperbaric hyperoxia is sufficient to inhibit the baroreceptor. As a neurotransmitter, NO is known to regulate baroreceptor activity through sympathetic afferent nerve fibers.28 Exogenous NO has an inhibitory action on the baroreceptor, albeit only at high concentrations, and under such circumstances that the baroreceptor activity recovers within seconds.29 Unlike classical NO activity, SNO-Hb bioactivity is mediated by species distinct from N·O itself (eg, low-mass SNOs).27,30,31 It is noteworthy that other endogenous S-nitrosothiols (eg, SNO-cysteine) can suppress baroreceptor activity independently of cGMP generation (the mediator of classical nitrosovasodilator activity).32 Stereoselective recognition sites in the baroreceptor vasculature could mediate baroreceptor inhibition by S-nitrosylated species such as SNO-cysteine and SNO-Hb.33 Expanded discussion is available online.
In conclusion, we have shown that SNO reconstitution of Hb successfully overcomes the reduction in tumor perfusion created by Hb itself. This activity involves the allosteric control of NO release by O2. The hemodynamic effects of SNO-Hb are a composite of central activity (baroreceptor inhibition) and allostery-facilitated NO release in the peripheral circulation. In contrast, native Hb, at concentrations characteristic of hemolytic states, has no central pressor effect. These results are summarized in Table 2.
Whereas the concentration of Hb we used is low from the O2 delivery standpoint, it greatly exceeds the concentration of any endogenous NO. In light of our results, development of safe Hb-based blood substitutes might include approaches that not only preserve Hb allostery (eg, crosslinking to prevent dissociation into dimmers) but also reconstitute SNO content. Reoxygenation of hypoxic tissues may benefit from both the central pressor and blood flow increasing effects elicited by SNO-Hb. Further, the treatment of trauma or septic patients might involve manipulation of SNO-Hb allostery to limit the excessive release of NO.
Acknowledgments
This work was supported by grants from the NIH/NCI CA40355 (M.W.D.), the Howard Hughes Medical Institute (J.S.S.), the NIH Division of Research Resources COBRE University of Puerto Rico Protein Research Center (Grant 5P20RR016439) (J.B.), and a collaborative travel award from the North Atlantic Treaty Organization (NATO) (G.T.). P.S. is a fellow of the Belgian American Educational Foundation (BAEF) and a ‘Fonds National de la Recherche Scientifique’ (FNRS) post-doctoral fellow detached from the University of Louvain Medical School (UCL, Unit of Pharmacology and Therapeutics, Brussels, Belgium). T.M. received support from the Institute for Medical Research, Inc, Durham Veteran’s Affairs Medical Center. The authors appreciate the assistance of Christine Baudelet and Pelzer O. Doar.
J.S.S. has a financial interest in Nitrox LLC, and is a paid consultant.
Both authors contributed equally to this work.
References
Jia L, Bonaventura C, Bonaventura J, Stamler JS. S-nitrosohaemoglobin: a dynamic activity of blood involved in vascular control. Nature. 1996; 380: 221eC226.
Stamler JS, Jia L, Eu JP, McMahon TJ, Demchenko IT, Bonaventura J, Gernert K, Piantadosi CA. Blood flow regulation by S-nitrosohemoglobin in the physiological oxygen gradient. Science. 1997; 276: 2034eC2037.
McMahon TJ, Moon RE, Luschinger BP, Carraway MS, Stone AE, Stolp BW, Gow AJ, Pawloski JR, Watke P, Singel DJ, Piantadosi CA, Stamler JS. Nitric oxide in the human respiratory cycle. Nat Med. 2002; 8: 711eC717.
Crawford JH, White CR, Patel RP. Vasoactivity of S-nitrosohemoglobin: role of oxygen, heme, and NO oxidation states. Blood. 2003; 101: 4408eC4415.
Gladwin MT, Lancaster JR Jr, Freeman BA, Schechter AN. Nitric oxide’s reactions with hemoglobin: a view through the SNO-storm. Nat Med. 2003; 9: 496eC500.
Gonzalez-Alonso J, Olsen DB, Saltin B. Erythrocyte and the regulation of human skeletal muscle blood flow and oxygen delivery: role of circulating ATP. Circ Res. 2002; 91: 1046eC1055.
Cosby K, Partovi KS, Crawford JH, Patel RP, Reiter CD, Martyr S, Yang BK, Waclawiw MA, Zalos G, Xu X, Huang KT, Shields H, Kim-Shapiro DB, Schechter AN, Cannon RO3rd, Gladwin MT. Nitrite reduction to nitric oxide by deoxyhemoglobin vasodilates the human circulation. Nat Med. 2003; 9: 1498eC1505.
James PE, Lang D, Tufnell-Barret T, Milsom AB, Frenneaux MP. Vasorelaxation by red blood cells and impairment in diabetes: reduced nitric oxide and oxygen delivery by glycated hemoglobin. Circ Res. 2004; 94: 976eC983.
Datta B, Tufnell-Barrett T, Bleasdale RA, Jones CJ, Beeton I, Paul V, Frenneaux M, James P. Red blood cell nitric oxide as an endocrine vasoregulator: a potential role in congestive heart failure. Circulation. 2004; 109: 1339eC1342.
Savitsky JP, Doczi J, Black J, Arnold JD. A clinical safety trial of stroma-free hemoglobin. Clin Pharmacol Ther. 1978; 23: 73eC80.
Bunn HF, Esham WT, Bull RW. The renal handling of hemoglobin. I. Glomerular filtration. J Exp Med. 1969; 129: 909eC923.
Alayash AI. Oxygen therapeutics: can we tame haemoglobin Nat Rev Drug Discov. 2004; 3: 152eC159.
Deem S, Kim JU, Manjula BN, Acharya AS, Kerr ME, Patel RP, Gladwin MT, Swenson ER. Effects of S-nitrosation and cross-linking of hemoglobin on hypoxic pulmonary vasoconstriction in isolated rat lungs. Circ Res. 2002; 91: 626eC632.
Doyle MP, Hoekstra JW. Oxidation of nitrogen oxides by bound dioxygen in hemoproteins. J Inorg Biochem. 1981; 14: 351eC358.
Eich RF, Li T, Lemon DD, Doherty DH, Curry SR, Aitken JF, Mathews AJ, Johnson KA, Smith RD, Phillips GNJR, Olson JS. Mechanism of NO-induced oxidation of myoglobin and hemoglobin. Biochemistry. 1996; 35: 6976eC6983.
Jourd’heuil D, Gray L, Grisham MB. S-nitrosothiol formation in blood of lipopolysaccharide-treated rats. Biochem Biophys Res Commun. 2000; 273: 22eC26.
Nakai K, Sakuma I, Togashi H, Yoshioka M, Sugawara T, Satoh H, Kitabatake A. S-nitrosylated polyethylene glycol-conjugated hemoglobin derivative as a candidate material for oxygen therapeutics. Adv Exp Med Biol. 2003; 519: 207eC216.
Hilf R, Michel I, Bell C, Freeman JJ, Borman A. Biochemical and morphologic properties of a new lactating mammary tumor line in the rat. Cancer Res. 1965; 25: 286eC299.
McMahon TJ, Stamler JS. Concerted nitric oxide/oxygen delivery by hemoglobin. Methods Enzymol. 1999; 301: 99eC114.
Dewhirst MW, Ong ET, Rosner GL, Rehmus SW, Shan S, Braun RD, Brizel DM, Secomb TW. Arteriolar oxygenation in tumour and subcutaneous arterioles: effects of inspired air oxygen content. Br J Cancer Suppl. 1996; 27: S241eCS246.
Sorg BS, Moeller BJ, Donovan O, Cao Y, Dewhirst MW. Hyperspectral imaging of hemoglobin saturation in tumor microvasculature and tumor hypoxia development. J Biomed Opt. 2005 (In press).
Dunn TJ, Braun RD, Rhemus WE, Rosner GL, Secomb TW, Tozer GM, Chaplin DJ, Dewhirst MW. The effects of hyperoxic and hypercarbic gases on tumour blood flow. Br J Cancer. 1999; 80: 117eC126.
Singel DJ, Stamler JS. Chemical physiology of blood flow regulation by red blood cells: the role of nitric oxide and S-nitrosohemoglobin. Annu Rev Physiol. 2005; 67: 99eC145.
Shan SQ, Rosner GL, Braun RD, Hahn J, Pearce C, Dewhirst MW. Effects of diethylamine/nitric oxide on blood perfusion and oxygenation in the R3230Ac mammary carcinoma. Br J Cancer. 1997; 76: 429eC437.
Hahn JS, Braun RD, Dewhirst MW, Shan S, Snyder SA, Taube JM, Ong ET, Rosner GL, Dodge RK, Bonaventura J, Bonaventura C, DeAngelo J, Meyer RE. Stroma-free human hemoglobin A decreases R3230Ac rat mammary adenocarcinoma blood flow and oxygen partial pressure. Radiat Res. 1997; 147: 185eC194.
Sonveaux P, Dessy C, Brouet A, Jordan BF, Gregoire V, Gallez B, Balligand JL, Feron O. Modulation of the tumor vasculature functionality by ionizing radiation accounts for tumor radiosensitization and promotes gene delivery. FASEB J. 2002; 16: 1979eC1981.
Lipton AJ, Johnson MA, Macdonald T, Lieberman MW, Gozal D, Gaston B. S-nitrosothiols signal the ventilatory response to hypoxia. Nature. 2001; 413: 171eC174.
Paton JF, Deuchars J, Ahmad Z, Wong LF, Murphy D, Kasparov S. Adenoviral vector demonstrates that angiotensin II-induced depression of the cardiac baroreflex is mediated by endothelial nitric oxide synthase in the nucleus tractus solitarii of the rat. J Physiol. 2001; 531 (Pt 2): 445eC58.
Harada S, Tokunaga S, Momohara M, Masaki H, Tagawa T, Imaizumi T, Takeshita A. Inhibition of nitric oxide formation in the nucleus tractus solitarius increases renal sympathetic nerve activity in rabbits. Circ Res. 1993; 72: 511eC516.
Pawloski JR, Swaminathan RV, Stamler JS. Cell-free and erythrocytic S-nitrosohemoglobin inhibits human platelet aggregation. Circulation. 1998; 97 (3): 263eC267.
Liu L, Yan Y, Zeng M, Zhang J, Hanes MA, Ahearn G, McMahon TJ, Dickfeld T, Marshall HE, Que LG, Stamler JS. Essential roles of S-nitrosothiols in vascular homeostasis and endotoxic shock. Cell. 2004; 116: 617eC628.
Matsuda T, Bates JN, Lewis SJ, Abboud FM, Chapleau MW. Modulation of baroreceptor activity by nitric oxide and S-nitrosocysteine. Circ Res. 1995; 76: 426eC433.
Davisson RL, Travis MD, Bates JN, Lewis SJ. Hemodynamic effects of L- and D-S-nitrosocysteine in the rat. Stereoselective S-nitrosothiol recognition sites. Circ Res. 1996; 79: 256eC262.(Pierre Sonveaux, Andrew M)