Olga Rafikova, Elena Sokolova, Ruslan Rafikov and Evgeny NudlerMechanisms and Clinical Implications

Control of Plasma Nitric Oxide Bioactivity by Perfluorocarbons : Physiological

Print ISSN: 0009-7322. Online ISSN: 1524-4539 Copyright © 2004 American Heart Association, Inc. All rights reserved.

is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231Circulation doi: 10.1161/01.CIR.0000148782.37563.F8

2004;110:3573-3580; originally published online November 22, 2004;Circulation. 

http://circ.ahajournals.org/content/110/23/3573World Wide Web at:

The online version of this article, along with updated information and services, is located on the

  http://circ.ahajournals.org//subscriptions/

is online at: Circulation Information about subscribing to Subscriptions: 

http://www.lww.com/reprints Information about reprints can be found online at: Reprints:

  document. Permissions and Rights Question and Answer this process is available in the

click Request Permissions in the middle column of the Web page under Services. Further information aboutOffice. Once the online version of the published article for which permission is being requested is located,

can be obtained via RightsLink, a service of the Copyright Clearance Center, not the EditorialCirculationin Requests for permissions to reproduce figures, tables, or portions of articles originally publishedPermissions:

by guest on November 5, 2012http://circ.ahajournals.org/Downloaded from

Control of Plasma Nitric Oxide Bioactivityby Perfluorocarbons

Physiological Mechanisms and Clinical Implications

Olga Rafikova, MD; Elena Sokolova, MSc; Ruslan Rafikov, MSc; Evgeny Nudler, PhD

Background—Perfluorocarbons (PFCs) are promising blood substitutes because of their chemical inertness andunparalleled ability to transport and upload O2 and CO2. Here, we report that PFC emulsions also efficiently absorb andtransport nitric oxide (NO).

Methods and Results—Accumulation of NO and O2 in PFC micelles results in rapid NO oxidation and generation ofreactive NOx species. Such micellar catalysis of NO oxidation leads to formation of vasoactive S-nitrosothiols (RSNO)in vitro and in vivo as detected electrochemically. The efficiency of PFC-mediated S-nitrosation depends on the amountof PFC in aqueous solution. The optimal PFC concentration that produced the maximum level of RSNO was �1%(vol/vol). Larger PFC amounts were progressively less efficient in generating RSNO and functioned simply as NO sink.These results explain the characteristic hemodynamic effects of PFCs. Intravenous bolus application of PFC (0.14 g/kg,�1% vol/vol) to Wistar-Kyoto rats decreased mean arterial pressure significantly (�10 mm Hg over 40 minutes).PFC-induced hypotension could be further stimulated (�17 mm Hg over 140 minutes) by exogenous thiols (cysteineand glutathione). In contrast, a larger amount of PFC (1 g/kg, �7% vol/vol) exhibited a strong hypertensive effect(11 mm Hg over 40 minutes).

Conclusions—The present study reveals a physiologically significant pool of endogenous plasma NO and underscores thecrucial role of the circulating hydrophobic phase in modulating its bioactivity. The results also establish PFC as aconceptually new pharmacological tool for various cardiovascular complications associated with NO imbalance.(Circulation. 2004;110:3573-3580.)

Key Words: blood pressure � hemodynamics � nitric oxide � oxygen � fluorocarbons

Nitric oxide (NO) is enzymatically produced by varioustypes of cells and represents a central mediator within

the cardiovascular system.1 As a free radical, NO is highlyunstable in vivo. Its primary targets include heme proteinssuch as guanylyl cyclase and free radical species, eg, O•�

2 .2

Another physiologically significant aspect of NO biochemis-try is the formation of thionitrite esters with cysteine (Cys)and Cys residues (S-nitrosothiols; RSNO). Small RSNOs (eg,S-nitrosoglutathione [GSNO]) and nitroso-derivatives of pro-teins such as albumin and hemoglobin are known to begenerated in vivo and exert NO-like activity. They inducevasodilation, inhibit platelet aggregation, and participate invarious signal transduction pathways.3–7 Because RSNOs arerelatively stable and can release NO when required onreaction with various reducing agents,8,9 they may serve as abuffering system that magnifies the range of NO action alongthe vascular tree.10,11

Although the pharmacological potential of RSNO is wellappreciated,12–14 the role of plasma NO and the mechanism ofendogenous RSNO formation under physiological conditions

are a source of considerable debate.15 The apparent limitationfor plasma NO to exert its biological function is the constantpresence of abundant hemoglobin (10 mmol/L) from redblood cells that converts NO into inactive metabolites atnear-diffusion–limited rates.16 Moreover, NO is unable toreact directly with thiols. It must be oxidized first to formnitrosative agents, such as N2O3. Because of the slow rate ofthe third-order reaction of NO with O2 (k �4�106 mol/L�2

per second),17,18 the formation of N2O3 and eventually RSNOshould depend primarily on the local concentration of NO andthe molecular environment. Indeed, strong acceleration ofNO oxidation occurs within lipid membranes19 and hydro-phobic pockets of plasma proteins20 that effectively sequesterNO from the aqueous phase. Recently, we provided experi-mental evidence suggesting that such micellar catalysis ofNO oxidation and N2O3 formation by serum albumin accountsfor the presence of vasoactive RSNO in the circulation underphysiological conditions.21

In the present study, we explored a synthetic hydrophobicphase (perfluorocarbon [PFC]) to manipulate plasma NO

Received May 21, 2004; revision received July 18, 2004; accepted August 2, 2004.From the Department of Biochemistry, New York University Medical Center, New York, NY.Correspondence to Dr Evgeny Nudler, New York University Medical Center, Department of Biochemistry, 550 First Ave, MSB #359, New York, NY

10016. E-mail evgeny.nudled@med.nyu.edu© 2004 American Heart Association, Inc.

Circulation is available at http://www.circulationaha.org DOI: 10.1161/01.CIR.0000148782.37563.F8

3573 by guest on November 5, 2012http://circ.ahajournals.org/Downloaded from

bioactivity and establish its physiological role. PFCs arehighly hydrophobic, inert chemicals that are not miscible withwater and are capable of dissolving large quantities of gases,including O2 and CO2. With sophisticated technology, it ispossible to generate stable PFC emulsions of small, uniformparticles that can be used as temporary intravascular oxygencarriers.22–24 Here, we show that by changing the volume andproperties of the circulating hydrophobic phase with PFCs,one can rapidly and precisely modulate NO transport anddelivery, thus controlling blood pressure, platelet aggrega-tion, and other NO-dependent processes. The results demon-strate the existence of a physiologically significant pool ofcirculating plasma NO and the crucial role of nitrosativechemistry in its vascular activity. We also suggest a concep-tually new pharmacological approach for treating disordersassociated with NO disparity.

MethodsChemicals and ReagentOne hundred milliliters of Perftoran emulsion (Perftoran Co) con-tains 13 g of perfluorodecalin (C10F18), 6.5 g of perfluoromethylcy-clohexylpiperidin (C12F23N), 4.0 g of Proxanol 268, 0.6 g of NaCl, 39mg of KCl, 19 mg of MgCl2, 65 mg of NaHCO3, 20 mg of NaH2PO4,0.2 g of glucose, and H2O. Physical properties of Perftoran includea particle size of 0.03 to 0.15 �mol/L, osmolality of 280 to 340mOsm, viscosity of 2.5 cP, pH 7.5, O2 solubility of 7.0 vol%(pO2�760 mm Hg, 20°C), and CO2 solubility of 60 vol/%(pCO2�760 mm Hg, 20°C). Saline in control experiments containedall of the chemicals listed above except PFCs and Proxanol. Otherchemicals were from Sigma unless specifically indicated.

Determination of NO Partitioning (Q)NO/H2O solution was prepared in an airtight device by bubbling NOgas (Aldrich) that had been purified from higher oxides by passing itthrough a 1-mol/L solution of KOH into water until the concentrationof dissolved NO reached 1 mmol/L. Water (Milli-Q grade) wasdeaerated by boiling and then cooling under argon (Praxair). The NOconcentration was measured immediately before the reaction with anISO-NO Mark II nitric oxide electrode (WPI). To determine QNO forPFC, the ice-cold Perftoran (10 mL) was degassed in the airtightdevice by bubbling with argon for 6 hours. Next, 1.3 mL of Perftoranwas added to 12.2 mL of NO solution (11.2 �mol/L) in the presenceof argon. The temperature was adjusted to 20°C, and the ISO-NOMark II electrode was inserted into the solution under argon andsealed. QNO is calculated as described in Figure 1.

Determination of Thiols and RSNOsTo prepare plasma samples (100 �L), 0.5 mL of arterial blood wascollected into heparinized plastic tubes and centrifuged at 4500 rpmat 4°C for 3 minutes. Plasma was processed immediately afterpreparation. Plasma GSH and Cys were detected by high-perfor-mance liquid chromatography (Waters 600E pump) at 230 nm witha multiwavelength detector (Waters 490E) and C18 column with amobile phase of 99% 0.1 mol/L monochloroacetate buffer pH 3 and1% methanol. In addition, thiols were detected with monobromo(trimethylammonio)bimane bromide as described previously.25 Thetotal amount of RSNO in vitro or in plasma was determinedelectrochemically with 1.5 mmol/L CuCl2 (Cu2�/Cu�) to rapidlydecompose all RSNO. The released NO was detected with theISO-NO Mark II electrode. GSNO and SNO-BSA were used asstandards. They were prepared by incubation of GSH or BSA with a2-fold molar excess of NaNO2 in acidified water on ice. To estimatethe fraction of low-molecular-weight (LMW) RSNO in total RSNO,the real-time kinetics of RSNO decomposition were traced asdescribed above except that 40 �mol/L CuCl2 was used (Figure 5).

AnimalsAdult male Wistar rats (300 to 400 g) were used. Both femoralarteries and the left femoral vein of anesthetized animals (urethane1.2 g/kg IP) were cannulated with polyethylene tubing (PE-50, WPI).An arterial catheter was used for continuous monitoring of aortablood pressure and heart rate and to withdraw blood samples. Thevenous catheter was used for drug administration. The same infusionrate of drugs or saline (0.2 mL/min) was applied for all animals. Thedrugs were used as follows: L-NAME 20 mg/kg, GSH 2 mg/kg, Cys0.8 mg/kg, NaNO2 1 mg/kg, and Perftoran 136 mg/kg (low dose) or1 g/kg (high dose). Before drug administration, mean arterialpressure (MAP) was recorded, and the initial level of RSNO wasdetermined in plasma as described above. MAP and heart rate weredetected with the pressure sensor connected to an amplifier. Thesignal was digitally converted at a PowerLab 200 workstation(ADInstruments) and transmitted to a desktop computer. All statis-tical calculations were performed with Origin 7.0 software (Origin-Lab Corp).

ResultsDramatic Acceleration of NO Oxidation by PFCTo study the effect of PFC on NO bioavailability, we chosePerftoran, the Russian brand of clinically approved bloodsubstitute.26 Perftoran contains 13% (wt/vol) perfluorodeca-line and 6.5% (wt/vol) perfluoromethylcyclopiperidine,which forms �0.07-�m micelles stabilized by a poloxamer-type surfactant (Figure 1). O2 solubility by Perftoran is �12

Figure 1. Micellar catalysis of NO oxidation by PFC. a, PFCcomponent of Perftoran emulsion. b, PFC as catalyst of NO oxi-dation and RSNO formation. Small and uniform PFC micelles ofPerftoran are stabilized by Proxanol surfactant. If partitioning (Q)for NO and O2 is �1, PFC micelle would increase local concen-tration of both gases by sequestering them from aqueous phaseand accelerate NO oxidation and N2O3 formation. c, Determina-tion of QNO for PFC. Equation used for QNO calculation: [NO]H2O

indicates NO concentration in aqueous phase (8.9 �mol/L);[NO]PFC, NO concentration in PFC micelle; VPFC, volume of PFC(1.3 mL); and �, change of [NO]H2O on addition of PFC. � wasdetermined experimentally. d, Sequestration of NO by PFC.Infusion of PFC into 12.2 mL of NO solution under O2-free argonconditions results in immediate decrease in [NO]H2O from 11.2(light gray bar) to 8.9 (dark gray bar) �mol/L as determined elec-trochemically (see Methods). Mean�SE (n�10; P0.01).

3574 Circulation December 7, 2004

by guest on November 5, 2012http://circ.ahajournals.org/Downloaded from

times greater than that in water or blood plasma.26 Unlike thechemical binding of O2 to the porphyrin-iron sites of hemo-globin, O2 dissolution in PFC is a simple, passive process inwhich hydrophobic gas molecules occupy PFC micelles. Incontrast to the characteristic sigmoid binding curve of O2 tohemoglobin, O2 solubility in PFC emulsions increases lin-early with partial pressure.22–24

Because NO is hydrophobic, we reasoned that it should beconcentrated by PFC micelles along with O2 under aerobicconditions, implying that PFC would accelerate NO oxidation(Figure 1b). To estimate the maximum acceleration value (H)of this reaction, the partition coefficient (QNO) for a PFC/H2Osystem was determined. To calculate QNO, which is describedby the equation shown in Figure 1c, we used a Clark-type NOelectrode attached to an airtight O2-free gas/liquid mixingdevice to monitor changes in NO concentration on addition ofPFC. We determined QNO(PFC/H2O) to be �200 (Figures 1cand 1d). QO2(PFC/H2O) has been shown to be �12.26 For thetrimolecular reaction of N2O3 formation, the equationH�Q2

NO�QO2�2002�12�5�105 is used. This estimation im-plies that even a small quantity of PFC in mammalian blood(eg, 1% vol/vol) would sequester a large amount of circulat-ing NO. At the same time, it should boost production ofreactive N2O3 species hundreds of thousands of times, whichin turn would result in increased nitrosation of externalmolecules such as thiols (Figure 1b).

Catalysis of RSNO Formation by PFC In VitroTo investigate the stimulating effect of PFC on RSNOformation, we compared the rate of GSNO accumulation as afunction of PFC volume in the probe. GSNO was detectedwith Cu2�/� followed by electrochemical detection of releasedNO.21 A representative experiment (Figure 2a) shows that 1%(vol/vol) PFC that has been aerobically saturated with watersolution of NO (0.5 �mol/L) increased the rate of GSNOformation �10-fold. A similar result was obtained withL-cysteine (data not shown). Importantly, the efficiency ofPFC-mediated S-nitrosation depends on the amount of PFC inthe probe (Figure 2b). The resonance-like maximum wasobserved at �1% (vol/vol) Perftoran. Larger or smalleramounts of PFC were progressively less efficient. This fact isreadily explained in terms of micellar catalysis of NOoxidation, in which a relatively small “optimal” volume of thehydrophobic phase (in this case, PFC) generates maximumacceleration of the reaction.21 These results suggest that theuse of various amounts of PFC can regulate the level ofendogenous vasoactive LMW RSNO.

PFC as a Powerful NO Sink In VivoTo evaluate the ability of PFC to absorb NO in vivo, westudied the effect of Perftoran on MAP in rats. Intravenousbolus application of PFC (1 g/kg, �7% vol/vol) was charac-terized by extensive hypertension (11 mm Hg) over the first40 minutes, followed by mild hypotension (Figure 3a).Because the PFC solution was isoosmotic (see Methods) andthe rate of its infusion was slow (0.2 mL/min), PFC-inducedvasoconstriction could not be attributed to the change in theosmotic status.

To confirm that the PFC-mediated increase in bloodpressure was due to plasma NO sequestration, we adminis-tered a nonselective NO synthase inhibitor, NG-nitro-L-arginine methyl ester (L-NAME; 50 mg/kg IP) to deplete thevasculature of endogenous NO. This caused a characteristicsteep increase in MAP (Figure 3b). Because PFC failed toincrease MAP any further in this case, we concluded thatintravascular NO was essential for PFC-inducedvasoconstriction.

To further substantiate the link between PFC-mediatedhypertension and plasma NO, we used sodium nitrite, awell-established NO source in vivo and a potent vasodilator.27

Reduction of nitrite to NO is catalyzed in vivo by xanthineoxidoreductase28 and deoxyhemoglobin29 and also can occurnonenzymatically.30 Moreover, a significant circulating arte-riovenous plasma nitrite gradient has been reported, whichargues in favor of intravascular nitrite being a stable naturalsource of NO.31,32 Infusion of NaNO2 (1 mg/kg IP) 90minutes after L-NAME decreased MAP sharply (Figure 3c).Subsequent PFC administration elevated MAP back to theoriginal L-NAME level (Figure 3c). Because nitrite is slowlyreduced to NO in vivo under physiological pH (hence itsvasorelaxation effect), an ability of PFC to specificallycounteract this effect confirms its mode of action as apowerful NO sink.

Taken together, these results demonstrate the existence ofa sizable pool of circulating NO in mammalian blood that

Figure 2. Catalysis of GSNO formation by PFC. a, Representa-tive NO-electrode tracing of Cu2�/�-dependent NO productionafter aerobic addition of 1 mmol/L GSH to 1 �mol/L NO solution(initial NO concentration was 100 �mol/L) in presence orabsence of PFC (1% vol/vol). b, PFC-mediated GSNO formationfunction of PFC concentration (experimental conditions as inFigure 2a; see Methods; n�6; P0.01).

Rafikova et al Control of Plasma NO Bioactivity by PFCs 3575

by guest on November 5, 2012http://circ.ahajournals.org/Downloaded from

serves as a systemic dilator. Limitation of its bioavailabilityby PFC leads to systemic vasoconstriction.

PFC-Mediated Catalysis of RSNOs In Vivo andControl of Blood PressureAs discussed above (Figure 2), PFCs not only limit theconcentration of free NO but also accelerate its oxidation andformation of RSNOs. Generation of vasoactive RSNOs canexplain the delayed dilation effects of PFC in the experimentsdescribed in Figure 3. A modest but reproducible hypotensionwas observed 60 minutes after PFC infusion (Figure 3a).Furthermore, in the nitrite-treated rats, MAP dropped belowthe baseline level 30 minutes after PFC infusion (Figure 3c),whereas in the control group (without PFC), MAP remained�25 mm Hg above baseline (Figure 3c). We interpret theseresults to mean that during the first phase of action, PFCsrapidly absorb plasma NO, which causes the initial increasein MAP; however, because local concentration of NO and O2

in PFC micelles becomes high, N2O3 forms at a much greaterrate, which leads to accumulation of vasoactive RSNOs.

To test whether PFCs can indeed decrease vascular tonevia generation of endogenous RSNOs, a series of in vivostudies that used �10 times less PFC than that used in theexperiments reported in Figure 3 were performed. The reasonfor using a smaller amount of PFC derives from the graph inFigure 2b, which shows that �1% PFC in aqueous solution isthe optimal amount that generates the maximum acceleration

of NO oxidation and RSNO formation. Therefore, one wouldexpect that at this concentration in vivo, PFC would convertplasma NO into vasoactive RSNO most efficiently. Indeed,the bolus infusion of the low dose of PFC (136 mg/kg IV, �1%vol/vol total blood) did not cause any increase in MAP (Figure4a). Instead, it produced a more rapid and pronounced decreasein MAP (�10 mm Hg over 40 minutes) than that observed withthe high dose of PFC (compare Figures 4a and 3a).

To confirm that the observed PFC-mediated vasodilationwas due to endogenous RSNO formation, we measured LMWRSNO in plasma before and after PFC administration (Figure4b). Plasma LMW RSNOs were determined as described inFigure 5, using a concentration of CuCl2 (40 �mol/L) selectedto decompose predominantly LMW RSNO. More stablehigh-molecular-weight (HMW) RSNOs (eg, S-NO-albumin)were largely resistant to such a low concentration of Cu2�/�

during the experiment (Figure 5). Using this approach, wefound that the basal level of LMW RSNO in rat plasma is10�6 nmol/L (n�6) and total RSNO is �200 nmol/L (Figure4). These numbers are in good agreement with the level ofRSNO previously detected in human plasma by a similarelectrochemical method.33,34 Because of RSNO rapid decom-position during sample preparation, the detected values ofRSNO reflect only a fraction of their actual amount in blood.This explains at least in part a wide range of RSNO values inthe circulation determined by different methods.34 As Figure4b shows, the level of plasma LMW RSNO increased

Figure 3. Relationship between PFC-mediated hemodynamic effects and NO. a, Effect of high dose of PFC on MAP of anesthetizedrats. Infusion of bolus 1 g/kg PFC (�7% vol/vol of total blood) provokes steep increase in blood pressure followed by gradual asymp-totic recovery and mild hypotension. Bars represent extremes of MAP during time course of continuous monitoring. Mean�SE (n�12,P0.05). �MAP indicates change in MAP. Arrow indicates time point of PFC or saline infusion. b, Inactivation of NOS abolishes hemo-dynamic effects of PFC. Infusion of L-NAME (50 mg/kg) causes prolonged and stable hypertension, which is not further affected by 1g/kg PFC. Mean�SE (n�8, P0.05). c, Delivery of NO donor resumes hemodynamic effects of PFC under L-NAME conditions. Infusionof NaNO2 1 mg/kg IP 110 minutes after L-NAME administration leads to partial recovery of MAP (�24 mm Hg). In this case, subse-quent infusion of PFC (1 g/kg) fully restored hypertensive effect of L-NAME (�MAP�28 mm Hg; n�10; P0.05). d, Schematic interpre-tation of experimental results. Artery is shown with endothelial cells producing NO that diffuses in all directions. In a, PFC micelles (graycircle) rapidly absorb plasma NO, thus causing vasoconstriction and hypertension. In b, NOS is inactivated, which limits the amount ofplasma NO and prevents further potential constriction by PFC. In c, reducing components of blood convert exogenous NO2

� to NO,thus causing partial relaxation from L-NAME. However, PFC micelles sequester exogenous NO and thus constrict vessel back toL-NAME position.

3576 Circulation December 7, 2004

by guest on November 5, 2012http://circ.ahajournals.org/Downloaded from

�5-fold 10 minutes after PFC injection (1% vol/vol) and thengradually decreased over the next hour. The burst of plasmaRSNO induced by PFC explains the drop in MAP in Figure4a. Significantly, despite the marked increase in the level ofLMW RSNO in response to PFC, the amount of total RSNOundergoes only a slight change, which indicates that PFC-mediated NO oxidation targets primarily the pool of LMWRSH in the circulation. Indeed, only small thiols are capableof penetrating PFC micelles efficiently. This result impliesthat LMW RSNO has a much greater specific vasoactivitythan the more abundant HMW RSNO. The latter can beconsidered as a relatively constant NO pool for storage and

transport purposes.10 A slight drop of total plasma RSNO 10minutes after PFC administration (the time of maximal LMWRSNO increase) reflects the process of PFC-mediated redis-tribution of NO from its HMW depot to actively used LMWRSNO.

We determined the concentration of the most common freeLMW thiols, GSH and L-CysH, in plasma of normally fedadult rats to be 15�3 and 39�6 �mol/L, respectively (seeMethods), which is consistent with previous measurements.35

Such a relatively low level of reduced small thiols in thecirculation could limit RSNO formation despite the high yieldof available N2O3 generated by PFC. To compensate for apotential deficiency of intravascular small thiols, we admin-istered exogenous LMW RSH to PFC-treated rats and exam-ined their effects on LMW RSNO formation and vasodilation.As Figure 6a demonstrates, coadministration of 2 mg/kg GSHalong with PFC resulted in a �68% increase in the amount ofplasma LMW RSNO (top panel) and a concomitant markeddecrease in MAP (bottom panel). Also, a strong effect on both

Figure 4. Effect of low dose of PFC on MAP and RSNO forma-tion in anesthetized rats. a, Systemic vasodilation response tointravenous bolus of 136 mg/kg PFC (1% vol/vol of total blood).Bars represent extremes of MAP during time course of continu-ous monitoring. Mean�SE (n�12, P0.05). Time of PFC orsaline infusion is indicated by arrow. �MAP indicates change inMAP. b, Stimulation of plasma LMW RSNO formation by PFC.�RSNO indicates change in plasma LMW RSNO. LMW RSNO inplasma was detected as described in Figure 5. Total RSNOconcentration in control (PFC) group was 210�25 (180�20),205�20 (160�15), and 195�25 (195�20) nmol/L at 0, 10, and25 minutes, respectively (n�10, P0.05). c, Schematic explain-ing vasodilating effect of PFC. At 1% vol/vol concentration, PFCmicelles (gray circle) absorb NO and O2 to generate reactiveN2O3 species via micellar catalysis of NO oxidation. N2O3 thennitrosates plasma small thiols to form vasoactive RSNOs thatcause vasorelaxation.

Figure 5. Method for quantitative discrimination of LMW RSNOfrom HMW RSNO. a, Electrochemical detection of GSNO inmixture with SNO-albumin. A representative NO-electrode trac-ing of real-time Cu2�/�-dependent NO production from probecontaining 200 nmol/L GSNO and 200 nmol/L BSA-NO in80 mmol/L Tris-HCl, pH 7.9. CuCl2 was added to 40 �mol/L attime 0. By 40 seconds, all GSNO but only �20% SNO-BSA hadbeen decomposed, which indicates marked difference between2 types of RSNO in their sensitivity to low dose of Cu2�/�.NO-Cys was more sensitive than GSNO and decomposedwithin first 15 seconds (data not shown). b, Calibration curves ofGSNO and BSA-NO decomposition. Plot represents NO releasedata from NO electrode at 40-second time point as function ofRSNO concentration. Under selected conditions (a), GSNO(slope 0.9, r�0.9998, P0.0001) is 90 times more sensitive toCuCl2 (40 �mol/L) than SNO-BSA (slope 0.01, r�0.96, P0.02).Such a difference allows estimation of amount of small RSNOsin mixture of total RSNO. Sensitivity of method is 5 nm/L, linear-ity 10 to 5000 nmol/L, recovery 95�5%.

Rafikova et al Control of Plasma NO Bioactivity by PFCs 3577

by guest on November 5, 2012http://circ.ahajournals.org/Downloaded from

RSNO formation (87% increase) and vasodilation(�17 mm Hg, 140 minutes) was observed when L-Cys (0.8mg/kg IV) was used instead of GSH (Figure 6b). WithoutPFC, neither glutathione nor cysteine had any effect onplasma RSNO or blood pressure at these concentrations.Obviously, the combined hemodynamic effect of PFC andthiols was much greater than that of PFC alone (compareFigures 4 and 6b), which provides direct evidence thatPFC-induced vasodilation was due to endogenous RSNOformation.

Because the half-life of Perftoran in human and animalblood is �24 hours,26 the eventual recovery of MAP after asingle dose of RSH was likely due to rapid RSH/RSNOuptake by tissues rather than PFC inactivation. The data ofFigure 6b confirm this conclusion. On receiving a seconddose of L-Cys (0.8 mg/kg) �2 hours after the first thiolinfusion, PFC-treated animals responded by a second wave ofmarked vasodilation that was virtually identical to the firstone. This result demonstrates that one can maintain the steadylevel of circulating RSNO and control blood pressure for along period of time simply by supplying small amounts ofnatural thiols to PFC-treated animals.

DiscussionThe physiological and pharmacological significance of thepresent study is several fold. First, it establishes the crucial

role of plasma NO in systemic hemodynamics. Second, itdirectly implicates the intravascular hydrophobic phase inregulation of plasma NO bioactivity via the micellar catalyticmechanism of endogenous RSNO formation. Third, it estab-lishes intrinsic plasma LMW RSNO as a potent relaxingfactor under noninflammatory conditions. Fourth, it explainsvarious hemodynamic effects associated with the use of PFCin the clinic. Fifth, it provides a conceptionally new pharma-cological approach to manipulate NO bioactivity.

Physiological Role of Plasma NO and theMechanism of Endogenous RSNO FormationIt has been widely assumed that a fraction of endothelium-derived NO that does not diffuse into the vessel wall getsrapidly trapped and inactivated by hemoglobin from redblood cells. Recently, we challenged this view by showingthat under physiological conditions, a significant amount ofNO accumulates in the hydrophobic compartments of serumalbumin.21 The high local concentration of NO within theprotein interior leads to accelerated NO oxidation and forma-tion of nitrosative N2O3 species, which in turn are responsiblefor maintaining the basal level of circulating vasoactiveRSNO.21 In support of this model, a number of studies,including the present one, detected the basal level of RSNO inmammalian plasma in the 20 to 200 nmol/L range.34 More-over, intravenous application of NO was shown to exert a

Figure 6. Cumulative effect of exogenous thiols and PFC on vascular tone and RSNO formation. a, Systemic vasodilation (top) andincrease in plasma RSNO (bottom) in response to intravenous bolus infusion of 136 mg/kg PFC and 2 mg/kg GSH. Bars representextremes of MAP during time course of continuous monitoring. Mean�SE from 11 independent experiments (n�11, P0.05). Arrowsindicate when chemicals were infused. �RSNO indicates change in plasma LMW RSNO. Total RSNOs in control (PFC�GSH) groupwas 210�25 (185�25), 205�20 (190�20), and 195�25 (220�25) nmol/L at 0, 10, and 25 minutes, respectively. b, Systemic vasodila-tion (top) and increase in plasma RSNOs (bottom) in response to intravenous bolus infusion of 136 mg/kg PFC and 0.8 mg/kg Cys.Mean�SE from 10 independent experiments (n�10, P0.05). Total RSNOs in control (PFC�Cys) group was 210�25 (200�20),205�20 (180�15), and 195�25 (230�20) nmol/L at 0, 10, and 25 minutes, respectively. c, Persistent hypotension in response to repet-itive administration of 0.8 mg/kg Cys followed by singe dose of 136 mg/kg PFC (n�6, P0.05). Arrows indicate when Cys was infused.d, Schematics explaining vasodilating effect of exogenous thiols in combination with PFC. Increased concentration of LMW RSH in ratvasculature leads to higher production of vasoactive RSNO via PFC-generated N2O3.

3578 Circulation December 7, 2004

by guest on November 5, 2012http://circ.ahajournals.org/Downloaded from

systemic hemodynamic effect via RSNO in humans.11 How-ever, it has remained unclear whether intrinsic plasma NO orRSNO played any significant role in cardiovascular pro-cesses. In the present study, we addressed this important issueby devising an approach to specifically modulate the pool ofplasma NO and RSNO without boosting NO production ordelivering any exogenous NO into the system.

The approach is based on unique physical and chemicalproperties of PFC emulsions.22–24 By utilizing a commercialPFC product, Perftoran (Figure 1a), we demonstrate that PFCmicelles adsorb NO from the aqueous phase, greatly acceler-ating NO oxidation and RSNO formation under aerobicconditions (Figures 1 and 2). In accord with the theory ofmicellar catalysis,21 the rate of NO oxidation and RSNOformation depends on the volume of the hydrophobic phase,with a resonance-like maximum at �1% vol/vol PFC (Figure2b). Larger or smaller amounts of PFC were progressivelyless effective. Next, we applied PFC as a tool for modulatingplasma NO bioavailability and RSNO formation in themammalian vasculature. Administration of a large dose ofPFC (7% vol/vol) caused extensive vasoconstriction due torapid sequestration of plasma NO (Figure 3). Alternatively, asmall dose of PFC (1% vol/vol) induced long-lasting sys-temic dilation (Figure 4). In this case, most of PFC-sequestered NO was efficiently converted into N2O3 to elicitendogenous vasoactive LMW RSNO (Figures 4 and 6).Because the basal level of endogenous LMW RSNO can beincreased dozens of fold simply by changing the property ofthe circulating hydrophobic phase, there must be a substantialamount of available NO in mammalian plasma at any givenmoment. Moreover, the level of plasma LMW RSNO can beelevated further by supplementing PFC with exogenousLMW RSH (Figure 6), which indicates that RSNO formationis limited by circulating thiols rather than by plasma NO.Importantly, the fate and physiological significance of plasmaNO are determined by the local hydrophobic environment.21

Taken together, the results of the present study directlyimplicate intrinsic plasma NO in maintaining vascular tonedespite the presence of red blood cells or other potential NOscavengers. Furthermore, our results underscore the physio-logical importance of nitrosative chemistry in plasma, estab-lishing endogenous RSNO as a significant transport anddelivery system of bioactive NO in the mammaliancirculation.

Pharmacological Applications of PFC as a PotentModulator of NO BioactivityPFC emulsions have been clinically evaluated in manycountries as artificial O2 carriers to reduce allogeneic bloodtransfusion and improve tissue oxygenation. They have beensuccessfully used to augment acute hemodilution after traumaor surgery and to treat conditions such as cerebral ormyocardial ischemia.22–24 Although PFCs are well toleratedin animals and humans, their unexpected pulmonary andsystemic hemodynamic effects have limited their clinicaluse.36,37 In particular, the higher doses of PFC have beenassociated with progressive hypertension and decreased heartrate.38 However, unexpected beneficial anti-inflammatoryand antithrombotic effects of PFCs have been noticed39,40 that

could not be explained merely by the ability of PFC totransport and upload O2 or change blood viscosity.

The present study establishes PFC as a potent modulator ofNO bioactivity, thus explaining, at least in part, its dose-dependent systemic effects. Most importantly, the presentfindings raise an attractive possibility of using PFC as apharmacological tool to influence vascular tone and plateletfunction by modulating plasma NO activity via endogenousblood-borne NO carriers. In this regard, PFCs represent aconceptually new drug for the treatment of cardiovasculardisorders associated with compromised and excessive NObioavailability. Indeed, depending on the dose, PFC functionseither as a potent NO stimulator or suppressor. The stimulat-ing effect occurs at lower doses owing to the boosting ofendogenous RSNO production (Figures 4 and 6). At higherdoses, PFC suppresses NO activity by decreasing its availableconcentration (Figure 3a).

Vasoregulation With Natural Thiols and PFC: ASafe Alternative to Exogenous RSNORSNOs have a wide variety of therapeutic effects and may beparticularly useful alternatives to organic nitrates becausethey do not elicit tolerance. As potent systemic vasodilatorsand platelet inhibitors, RSNOs have been evaluated in thetreatment of various cardiovascular complications, includingthrombosis, vasospasm, preeclampsia, and ischemia/reperfu-sion injury.12–14 Furthermore, because of their bronchodilat-ing properties and the fact that the airway level of RSNOs isreduced in asthma and cystic fibrosis patients, RSNOs areconsidered as promising therapeutics for pulmonary condi-tions.41,42 Finally, RSNOs may be used as antiviral andantimicrobial agents, because they inhibit viral proteaseenzymes and possess a strong bacteriostatic effect.43,44 How-ever, despite growing interest in the pharmacological poten-tial of RSNOs, a number of general problems limit theirclinical use. The main shortcoming of RSNOs as a potentialdrug is an unpredictable rate of decomposition in vivo andinstability in physiological vehicles. Excess RSNOs can leadto hypotension and hemorrhage due to impaired plateletfunction. Furthermore, synthetic drugs that have beenS-nitrosylated retain their own side effects.

In this regard, the use of PFC and PFC/RSH combinationsprovides a unique pharmacological solution for safe deliveryof RSNOs. Because PFC-mediated generation of RSNOsoccurs in vivo and can be precisely controlled by PFC andexogenous RSH concentrations, concerns about RSNO over-dose and storage problems are eliminated. Furthermore,chemically inert PFC emulsions (eg, clinically safe Perftoran)and natural thiols, such as L-Cys or GSH, should haveminimal immunologic side effects. Finally, because PFCuncouples RSNO formation from overall NO synthesis, itcreates a unique opportunity to relieve nitrosative stress, ie,excessive nitration of macromolecules and tissue damageduring inflammation,45,46 while maintaining control overvascular tone, blood clotting, and organ perfusion.

AcknowledgmentsThese studies were supported by the Irma T. Hirschl Career Awardand a grant from the National Institutes of Health.

Rafikova et al Control of Plasma NO Bioactivity by PFCs 3579

by guest on November 5, 2012http://circ.ahajournals.org/Downloaded from

References1. Moncada S. Nitric oxide in the vasculature: physiology and pathophys-

iology. Ann N Y Acad Sci. 1997;811:60–67.2. Hanafy KA, Krumenacker JS, Murad F. NO, nitrotyrosine, and cyclic

GMP in signal transduction. Med Sci Monit. 2001;7:801–819.3. Ignarro LJ, Lippton H, Edwards JC, et al. Mechanism of vascular smooth

muscle relaxation by organic nitrates, nitrites, nitroprusside, and nitricoxide: evidence for the involvement of S-nitrosothiols as active interme-diates. J Pharmacol Exp Ther. 1981;218:739–749.

4. Simon DI, Stamler JS, Jaraki O, et al. Antiplatelet properties of proteinS-nitrosothiols derived from nitric oxide and endothelium-derivedrelaxing factor. Arterioscler Thromb. 1993;13:791–799.

5. Scharfstein JS, Keaney JF Jr, Slivka A, et al. In vivo transfer of nitricoxide between a plasma protein-bound reservoir and low molecularweight thiols. J Clin Invest. 1994;94:1432–1439.

6. Jia L, Bonaventura C, Bonaventura J, et al. S-nitrosohaemoglobin: adynamic activity of blood involved in vascular control. Nature. 1996;380:221–226.

7. Stamler JS, Lamas S, Fang FC. Nitrosylation: the prototypic redox-basedsignaling mechanism. Cell. 2001;106:675–683.

8. Singh RJ, Hogg N, Joseph J, et al. Mechanism of nitric oxide release fromS-nitrosothiols. J Biol Chem. 1996;271:18596–18603.

9. Kashiba-Iwatsuki M, Kitoh K, Kasahara E, et al. Ascorbic acid andreducing agents regulate the fates and functions of S-nitrosothiols.J Biochem. 1997;122:1208–1214.

10. Jourd’heuil D, Hallen K, Feelisch M, et al. Dynamic state ofS-nitrosothiols in human plasma and whole blood. Free Radic Biol Med.2000;28:409–417.

11. Rassaf T, Kleinbongard P, Preik M, et al. Plasma nitrosothiols contributeto the systemic vasodilator effects of intravenously applied NO: experi-mental and clinical study on the fate of NO in human blood. Circ Res.2002;91:470–477.

12. Hogg N. Biological chemistry and clinical potential of S-nitrosothiols.Free Radic Biol Med. 2000;28:1478–1486.

13. Al-Sa’doni H, Ferro A. S-Nitrosothiols: a class of nitric oxide-donordrugs. Clin Sci (Lond). 2000;98:507–520.

14. Richardson G, Benjamin N. Potential therapeutic uses for S-nitrosothiols.Clin Sci (Lond). 2002;102:99–105.

15. Gladwin MT, Lancaster JR Jr, Freeman BA, et al. Nitric oxide’s reactionswith hemoglobin: a view through the SNO-storm. Nat Med. 2003;9:496–500.

16. Joshi MS, Ferguson TB Jr, Han TH, et al. Nitric oxide is consumed, ratherthan conserved, by reaction with oxyhemoglobin under physiologicalconditions. Proc Natl Acad Sci U S A. 2002;99:10341–10346.

17. Wink DA, Nims RW, Darbyshire JF, et al. Reaction kinetics for nitro-sation of cysteine and glutathione in aerobic nitric oxide solutions atneutral pH: insights into the fate and physiological effects of interme-diates generated in the NO/O2 reaction. Chem Res Toxicol. 1994;7:519–525.

18. Lewis RS, Deen WM. Kinetics of the reaction of nitric oxide with oxygenin aqueous solutions. Chem Res Toxicol. 1994;7:568–574.

19. Liu X, Miller MJ, Joshi MS, et al. Accelerated reaction of nitric oxidewith O2 within the hydrophobic interior of biological membranes. ProcNatl Acad Sci U S A. 1998;95:2175–2179.

20. Nedospasov A, Rafikov R, Beda N, et al. An autocatalytic mechanism ofprotein nitrosylation. Proc Natl Acad Sci U S A. 2000;97:13543–13548.

21. Rafikova O, Rafikov R, Nudler E. Catalysis of S-nitrosothiols formationby serum albumin: the mechanism and implication in vascular control.Proc Natl Acad Sci U S A. 2002;99:5913–5918.

22. Lowe KC. Perfluorinated blood substitutes and artificial oxygen carriers.Blood Rev. 1999;13:171–184.

23. Lane TA. Perfluorochemical-based artificial oxygen carrying red cellsubstitutes. Transfus Sci. 1995;16:19–31.

24. Spahn DR. Blood substitutes: artificial oxygen carriers: perfluorocarbonemulsions. Crit Care. 1999;3:93–97.

25. Velury S, Howell SB. Measurement of plasma thiols after derivatizationwith monobromobimane. J Chromatogr. 1988;424:141–146.

26. Gervits LL. Perfluorocarbon-based blood substitutes: Russian experience.In: Fluorine in Medicine in 21st Century. Banks RE, Lowe KC, eds.Shrewsbury, UK: Rapra; 1994.

27. Dejam A, Hunter CJ, Schechter AN, et al. Emerging role of nitrite inhuman biology. Blood Cells Mol Dis. 2004;32:423–429.

28. Godber BL, Doel JJ, Sapkota GP, et al. Reduction of nitrite to nitric oxidecatalyzed by xanthine oxidoreductase. J Biol Chem. 2000;275:7757–7763.

29. Cosby K, Partovi KS, Crawford JH, et al. Nitrite reduction to nitric oxideby deoxyhemoglobin vasodilates the human circulation. Nat Med. 2003;9:1498–1505.

30. Zweier JL, Samouilov A, Kuppusamy P. Non-enzymatic nitric oxidesynthesis in biological systems. Biochim Biophys Acta. 1999;1411:250–262.

31. Gladwin MT, Shelhamer JH, Schechter AN, et al. Role of circulatingnitrite and S-nitrosohemoglobin in the regulation of regional blood flowin humans. Proc Natl Acad Sci U S A. 2000;97:11482–11487.

32. Rassaf T, Feelisch M, Kelm M. Circulating NO pool: assessment of nitriteand nitroso species in blood and tissues. Free Radic Biol Med. 2004;36:413–422.

33. Goldman RK, et al. Nitrosothiol quantification in human plasma. AnalBiochem.1998;259:98–103.

34. Giustarini D, Milzani A, Colombo R, et al. Nitric oxide, S-nitrosothiolsand hemoglobin: is methodology the key? Trends Pharm Sci. 2004;25:311–316.

35. Ookhtens M, Mittur AV, Erhart NA. Changes in plasma glutathioneconcentrations, turnover, and disposal in developing rats. Am J Physiol.1994;266:979–988.

36. Faithfull NS, King CE, Cain SM. Peripheral vascular responses to fluo-rocarbon administration. Microvasc Res. 1987;33:183–193.

37. Mattrey RF, Hilpert PL, Long CD, et al. Hemodynamic effects of intra-venous lecithin-based perfluorocarbon emulsions in dogs. Crit Care Med.1989;17:652–656.

38. Johnson EC, Erickson BK, Podolsky A, et al. Effects of a perfluorocarbonemulsion for enhanced O2 solubility on hemodynamics and O2 transportin dogs. J Appl Physiol. 1995;79:1777–1786.

39. Lowe KC. Second-generation perfluorocarbon emulsion blood sub-stitutes. Artif Cells Blood Substit Immobil Biotechnol. 2000;28:25–38.

40. Heard SO, Puyana JC. The anti-inflammatory effects of perfluorocarbons:let’s get physical. Crit Care Med. 2000;28:1241–1242.

41. Gaston B, Sears S, Woods J, et al. Bronchodilator S-nitrosothiol defi-ciency in asthmatic respiratory failure. Lancet. 1998;351:1317–1319.

42. Foster MW, McMahon TJ, Stamler JS. S-nitrosylation in health anddisease. Trends Mol Med. 2003;9:160–168.

43. Sehajpal PK, Basu A, Ogiste JS, et al. Reversible S-nitrosation andinhibition of HIV-1 protease. Biochemistry. 1999;38:13407–13413.

44. De Groote MA, Testerman T, Xu Y, et al. Homocysteine antagonism ofnitric oxide-related cytostasis in Salmonella typhimurium. Science. 1996;272:414–417.

45. Beckman JS, Koppenol WH. Nitric oxide, superoxide, and peroxynitrite:the good, the bad, and ugly. Am J Physiol. 1996;271:1424–1437.

46. Nakazawa H, Fukuyama N, Takizawa S, et al. Nitrotyrosine formationand its role in various pathological conditions. Free Radic Res. 2000;33:771–784.

3580 Circulation December 7, 2004

by guest on November 5, 2012http://circ.ahajournals.org/Downloaded from