NITRIC OXIDE REGULATION OF NA+/K+-ATPASE ACTIVITY 275

Copyright © 2003 John Wiley & Sons, Ltd. J. Appl. Toxicol. 23, 275–278 (2003)

Insights into the Mechanism of ErythrocyteNa+++++/K+++++-ATPase Inhibition by Nitric Oxideand Peroxynitrite Anion

Pablo Muriel,1,* Gilberto Castañeda,1 Mónica Ortega1 and François Noël2

1 Sección Externa de Farmacología, CINVESTAV-I.P.N., Apdo. Postal 14-740, México 07000, D. F. México2 Departamento de Farmacología Básica e Clínica, Instituto de Ciências Biomédicas, Universidade federal do Rio deJaneiro, Cidade Universitária, 21941-590 Rio de Janeiro, Brazil

Key words: nitric oxide; Na+/K+-ATPase; SIN-1; SNAP; erythrocyte membranes.

Evidence shows that Na+++++/K+++++-ATPase from kidney, brain and liver is inhibited by nitric oxide (NO) andperoxynitrite anion (ONO−−−−−

2), but the mechanism is unknown. The aim of the present work was to study theinhibitory effect of NO and ONO−−−−−

2 on erythrocyte Na+++++/K+++++-ATPase. Erythrocyte membranes were isolated frommale Wistar rats by hypotonic washing. The membranes (free from haemoglobin) were incubated for Na+++++/K+++++-ATPase activity measurement at various concentrations of ATP in the presence or absence of 400 µµµµµM SNAP(an NO donor) or 100 µµµµµM SIN-1 (an ONO−−−−−

2 donor). At these concentrations, SNAP and SIN-1 released aboutthe same amount (100 µµµµµM) of NO or ONO−−−−−

2, respectively, as monitored by measuring NO−−−−−2 + NO−−−−−

3. Both SNAPand SIN-1 decreased Vmax by ca. 75% but they did not decrease the apparent affinity of the Na+++++/K+++++-ATPasefor the substrate (a decrease of Km was even observed after SNAP treatment). The pattern of this inhibitionis compatible either with oxidation of SH groups directly involved in ATP binding but in a way that is notsurmountable by increasing the substrate concentration (‘non-competitive’) or with oxidation of SH groupslocated outside the active site of the enzyme but important for the activity of the enzyme. Copyright © 2003John Wiley & Sons, Ltd.

* Correspondence to: P. Muriel, Pharmacology Section, CINVESTAV-I.P.N., Apdo. Postal 14-740, México 07000, D. F. México.E-mail: pamuriel@mail.cinvestav.mxContract/grant sponsor: CONACYT; Contract/grant number: 34394M.

JOURNAL OF APPLIED TOXICOLOGYJ. Appl. Toxicol. 23, 275–278 (2003)Published online in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/jat.922

Received 31 August 2002Revised 30 January 2003

Copyright © 2003 John Wiley & Sons, Ltd. Accepted 30 March 2003

INTRODUCTION

During the past decade, particular attention has been paidto the small, diffusible, unique molecule of nitric oxide(NO). Nitric oxide, the end product of the enzyme NOsynthase, influences various physiological processes inessentially every organ and tissue. It has a remarkablybroad spectrum of functions such as regulation of vasculartone, neurotransmission, antimicrobial defence mechanismsand immunomodulation. It is also involved in normal regul-ation of membrane function, particularly of Na+/K+-ATPaseactivity (Muriel and Sandoval, 2000a). However, themechanism by which NO modulates the activity of Na+/K+-ATPase is not fully understood.

Evidence indicates that NO-generating compoundsinhibit the activity of Na+/K+-ATPase and other mem-brane P-type ATPases. Sato et al. (1997) found that NO-generating compounds inhibit purified Na+/K+-ATPasefrom porcine cerebral cortex, and Boldyrev et al. (1997)reported that thiol-containing NO derivatives inhibit theactivity of bovine brain and dog kidney Na+/K+-ATPase.Recently, we found that NO and ONO−

2-generating com-pounds inhibit the Ca2+-ATPase activity present in hepatic

basolateral membranes (Muriel and Sandoval, 2000b).Because the function of membrane proteins is modulatedby the composition and fluidity of the lipid bilayer(Kimelberg and Papahadjopoulos, 1974; Muriel andMourelle, 1990), we previously studied if NO or ONO−

2

inhibits ATPase activity by altering the fluidity of themembrane in which it is embedded (Muriel and Sandoval,2000a). Although both NO and ONO−

2 inhibited Na+/K+-ATPase activity, the former increased the membranefluidity and the latter decreased it (Muriel and Sandoval,2000a), suggesting that alteration of membrane fluidity isnot the mechanism involved in ATPase inhibition.

Nitric oxide is a short-lived free radical that may inter-act with reactive oxygen intermediates to form more toxicspecies. The reaction of NO with superoxide anion pro-duces ONO−

2, which can decompose to generate a strongoxidant such as the hydroxyl radical (Beckman et al., 1990).As a consequence, it seems likely that NO and ONO−

2

decrease ATPase activity by oxidation of the sulphydrylgroups of the enzyme that are so important for its activity(Miller and Farley, 1990). This hypothesis is further sup-ported by the fact that ONO−

2 itself can induce sulphydryloxidation (Radi et al., 1991).

In an attempt to contribute to the elucidation of raterythrocyte Na+/K+-ATPase inhibition by NO and ONO−

2,we studied the kinetics of ATP activation after exposi-tion to SNAP and SIN-1, which are specific donors ofNO (Shaffer et al., 1992) and ONO−

2 (Groot et al., 1993),respectively.

276 P. MURIEL ET AL.

Copyright © 2003 John Wiley & Sons, Ltd. J. Appl. Toxicol. 23, 275–278 (2003)

MATERIALS AND METHODS

Preparation of erythrocyte membranes

The isolation of erythrocyte plasma membranes was per-formed according to the method of Rose and Oklander(1965) but with some modifications. Routinely, eightmale Wistar rats (250–300 g) fed Purina chow diet adlibitum were sacrificed under pentobarbital anaesthesia(25 mg kg−1, i.p.) by exsanguination through cardiac punc-ture. Blood samples (10 ml) containing heparin (50 IU)were centrifuged for 15 min at 2000 rpm (950 g) in the cold,and the plasma and buffy coat were aspirated. The eryth-rocytes were resuspended in 5–10 volumes of ice-coldisotonic saline and centrifuged as above, a process thatwas repeated three times. The erythrocytes obtained weretransferred into 30 ml of hypotonic medium (10 mMTRIS·HCl, pH 7.4, containing 0.1 mM EGTA). Thetubes were shaken vigorously and centrifuged at 4 °C for20 min at 20 000 rpm (48 000 g). After aspiration of thesupernatant the cells were resuspended in hypotonicmedium and treated as above, several times, until the lysedmembranes were free from haemoglobin. The final pelletof erythrocyte membranes was resuspended in distilledwater and subjected to three cycles of freezing and thaw-ing in liquid nitrogen at 37 °C. The final preparations werestored at −70 °C until assay. All animals received humancare and the study complies with the institution’s guide-lines and the Mexican Official Regulation (NOM-062-Z00-1999) regarding technical specifications for the production,care and use of laboratory animals.

Exposure of erythrocyte membranes to nitric oxideand peroxynitrite anion

A pool of erythrocyte membranes obtained from eight ratswas used for the determination of Na+/K+-ATPase activity.The membranes were divided into two groups. Aliquots ofthe first group (8 ml, 100 µg protein ml−1) were exposed toSNAP at a concentration of 400 µM in 20 mM imidazole(pH 7.4) for 2 h at 37 °C in a shaking bath. Similarly,aliquots of the second group were exposed to 100 µMSIN-1. The concentrations of SNAP and SIN-1 werechosen from previous observations (Muriel and Sandoval,2000a,b). After incubation, the Na+/K+-ATPase activity wasmeasured as described below. All the experiments wereperformed in duplicate and repeated three times withthree different pools of erythrocyte membranes. Proteindeterminations were performed according to the methoddescribed by Bradford (1976) using serum bovine albuminas the standard.

Enzymatic assay

The standard medium for total ATPase activity contained,in 0.5 ml, 90 mM NaCl, 10 mM KCl, 54 mM MgCl2, 20 µgof membrane protein and 20 mM imidazole buffer. Theconcentration of Na2ATP varied from 0.20 to 2.08 mM, sothat only the low-affinity/high-capacity sites for ATP wereexplored (Robinson, 1976). The pH of the solution wasadjusted to 7.4 at 37 °C, which is the temperature of allenzymatic incubations. The Mg2+-activated ATPase (basalMg2+-ATPase) was measured in an identical medium butcontaining 1 mM ouabain in order to inhibit fully the

α1-isoform of rat (Noël and Godfraind, 1984), which is theprincipal Na+/K+-ATPase isoform present in rat erythro-cytes (Post et al., 1960, Chang et al., 1998). The differencebetween total and Mg2+-ATPase represented the Na+/K+-ATPase. Incubations were carried out at 37 °C for 30 min.To stop the reaction, 100 µl of ice-cold 35% (w/v)trichloroacetic acid were added. Aliquots of 20 µl wereanalysed for inorganic phosphate according to Ames(1966). The spontaneous hydrolysis of ATP was monitoredby eliminating protein from control assays.

Nitrite +++++ nitrate determination

The NO−2 + NO−

3 levels in all the incubations were deter-mined by a method based on the Gries reaction (Greenet al., 1982). Briefly, 400 µl of incubation media were cen-trifuged at 2000 g with a micropore filter (ultrafree MCmicrocentrifuge device, UFC 3; Millipore, Bedford, MA)to remove substances heavier than 10 kDa. After passingthe samples through a copper-plated cadmium column fornitrate reduction, nitrite was measured by its absorbanceat 540 mm after mixing with a reagent consisting of0.2% naphthylethylenediamine dihydrochloride and 0.4%procaine in 6% trichloroacetic acid. The efficiency ofcadmium in converting nitrate to nitrite was verified bymeasuring both nitrate and nitrite standards before andafter sample measurement. The value obtained expressesthe total amount of serum NO or peroxynitrite end pro-ducts, namely NO−

2 + NO−3.

Statistical analysis

The ATP activation of Na+/K+-ATPase was representedgraphically using the classical concentration–activity plot.On the other hand, Km and Vmax were calculated using theclassical model of Michaelis-Menten (one activation site,no cooperativity) and a computerized non-linear regres-sion technique that adjusts the parameters to minimize thesum of absolute squared errors. The software used wasGraphPad Prism, version 1.03 (GraphPad Software Inc.San Diego, CA, USA). Experimental data are expressedas means ± SEM and compared by the use of ANOVA,followed by Newman–Keuls test to detect statistical differ-ences. Statistical significance was considered when P < 0.05.

RESULTS

Figure 1 shows that in control conditions, without SNAPor SIN-1, the media and erythrocyte membranes were freefrom NO−

2 and NO−3. Both SNAP (400 µM) and SIN-1

(100 µM) released the same amount of NO and ONO−2,

respectively, because they liberated the same amount ofend products, providing ca. 97 µM (NO−

2 + NO−3) (Fig. 1) —

a concentration that produced a clear and significant, butnot total, inhibition of hepatic Na+/K+-ATPase activity(Muriel and Sandoval, 2000a).

As a consequence, these concentrations of SNAP andSIN-1 were selected to test the effect of equimolar con-centrations of NO and ONO−

2 on erythrocyte NA+/K+-ATPase at various concentrations of the substrate (ATP).

As expected for the range of ATP concentrations used,the activity of the enzyme increased with ATP accordingto Michaelis–Menten kinetics (Fig. 2). Both SNAP and

NITRIC OXIDE REGULATION OF NA+/K+-ATPASE ACTIVITY 277

Copyright © 2003 John Wiley & Sons, Ltd. J. Appl. Toxicol. 23, 275–278 (2003)

SIN-1 decreased the maximal velocity of the enzyme byca. 75% but not the apparent affinity for the substrate(Table 1). In fact, a decrease of Km, rather than an in-crease, was observed with SNAP (Table 1).

DISCUSSION

Our results show that both NO and ONO−2 released by

SNAP and SIN-1, respectively, inhibit the rat erythrocyteNa+/K+-ATPase. At equimolar concentrations of NO andONO−

2 liberated, the two compounds were equieffectivein reducing (by ca. 75%) the Vmax of the enzyme. Usingthe same concentration of SIN-1, Sato et al. (1997) re-ported a similar degree of inhibition (80%) at a nearlysaturating concentration of ATP (2 mM). On the otherhand, these authors reported that SNAP, even at a higherconcentration (1 mM) than that used here (400 µM), wasless effective (ca. 25% inhibition) under the same con-ditions. A possible explanation for this discrepancy couldbe ascribed to the different isoforms present in porcinecerebral cortex (essentially α2 + α3) and rat erythrocyte(essentially α1), even if Na+/K+-ATPase from brain hasbeen reported to be more vulnerable to oxidative modific-ation than that from kidney (Xie et al., 1995; Boldyrevet al., 1997).

In searching for new insights into the molecular mechan-ism of inhibition, we performed classical activation curvesfor the substrate ATP in the presence of a fixed concen-tration of the inhibitors. Our data show that SNAP andSIN-1 reduce Vmax with no diminution of the apparentaffinity for the substrate (in fact, a decrease, and notincrease, of Km has to be considered after SNAP treat-ment). Classically, this pattern of inhibition indicates a non-competitive inhibition, a descriptive term that is supportedby different molecular mechanisms but is compatible witha decrease in the number of (functional) enzymes. Such asituation is observed after phosphorylation of the acetyl-cholinesterase esteratic site by organophosphates for inst-ance (Kardos and Sultatos, 2000). Note that this so-called‘irreversible’ inhibition can be reversed by nucleophilicagents, such as pralidoxime.

Another possibility is the action at an allosteric site,distinct from the catalytic site, leading to a decrease ofenzyme efficacy. In the present case, an oxidation ofsulphydryl groups essential for ATPase activity is an at-tractive possibility. In fact, different experiments argue forsuch a mechanism, beginning with the fact that the reduc-tion of disulphide bridges (at the α- and/or β-subunit) hasbeen claimed to inhibit Na+/K+-ATPase activity (Miller andFarley, 1990). Sato et al. (1997) found that NO-generatingcompounds (including SNAP and SIN-1) inhibit purifiedNa+/K+-ATPase activity from porcine cerebral cortex, prob-ably due to oxidation of the SH group. Boldyrev et al.(1997) found that thiol-containing NO derivatives inhibitthe activity of brain and kidney Na+/K+-ATPase, accompa-nied by a decrease in the number of protein thiol groups.In addition, dithiothreitol (DTT, a thiol-group-protectingagent) prevents liver ATPase inhibition (Muriel andSandoval, 2000a) and can partly reactivate the oxidizedstate of the enzyme (Sato et al., 1997).

Finally, restoration of Na+/K+-ATPase activity by DTTis accompanied by restoration of SH groups of the enzyme(Boldyrev et al., 1997). However, the location and natureof the sulphydryl groups involved continue to be an openquestion. On the one hand, the present results showing adecrease of Vmax without an increase of Km could indicatethat oxidation affects sulphydryl groups participating inthe ATP cleavage at the active cleft, in a manner similar tothat described above for organophosphate inhibition of

Figure 1. Effect of SNAP, a nitric oxide (NO) donor, and SIN-1, aperoxynitrite (ONO−

2) donor, on the liberation of end products(expressed as NO−

2 + NO−3) during incubation of erythrocyte

membranes. Each bar represents the mean ± SEM of three ex-periments performed in triplicate.

Figure 2. The Na+/K+-ATPase activity in erythrocyte membranesexposed to SNAP (400 µM) and SIN-1 (100 µM). Each point repre-sents the mean ± SEM from three independent experiments.Curves were drawn according to the parameters calculated bynon-linear regression analysis of the data (see values in Table 1).

Table 1—The Vmax and Km valuesa of rat erythrocyte Na+/K+-ATPase in the absence (control) or presence of either SNAP(400 µM) or SIN-1 (100 µM)

Treatment Vmax Km

Control 2.000 ± 0.096 0.694 ± 0.083SNAP 0.568 ± 0.047b 0.234 ± 0.078b

SIN-1 0.511 ± 0.049b 0.413 ± 0.122

a Parameters were estimated by non-linear regression analysisof the data depicted in Fig. 1. Values correspond to the fittedparameters for the mean curves constructed from three experi-ments ± standard error.b Means different from control group; Student’s t-test (P < 0.005).

278 P. MURIEL ET AL.

Copyright © 2003 John Wiley & Sons, Ltd. J. Appl. Toxicol. 23, 275–278 (2003)

acetylcholinesterase. As an alternative to this first hypo-thesis, we cannot rule out a putative oxidation of SH groupslocated into the active site but not participating directly inthe binding of ATP, so that there is no real competition atthe very same binding site. On the other hand, importantSH groups, independently of their localization, could beimplicated in the folding of this enzyme (Hu et al., 2000).

In conclusion, the present report supports the view thatNO and ONO−

2 donors inhibit erythrocyte Na+/K+-ATPasethrough the production of oxidative end products that

interact reversibly with sulphydryl groups probably locatedoutside the catalytic site of the enzyme or that do not part-icipate in substrate binding if they are in the catalytic cleft.

Acknowledgements

The authors express their gratitude to Ms Patricia González, Mr Mario G.Moreno and M. Ramón Hernández for their excellent technical assistance andMr Alfredo Padilla for preparing the figures. This work was supported in partby CONACYT grant 34394M.

REFERENCES

Ames BN. 1966. Assay of inorganic phosphate, total phosphateand phosphatases. Methods Enzymol. 8: 115–118.

Beckman JS, Beckman TW, Chen J, Marshall PA, Freeman BA.1990. Apparent hydroxyl radical production by peroxynitrite:implications for endothelial injury from nitric oxide andsuperoxide. Proc. Natl. Acad. Sci. USA 87: 1620–1624.

Boldyrev AA, Bulygina ER, Kramarenko GG, Vanin AF. 1997.Effect of nitroso compounds on Na/K-ATPase. Biochim.Biophys. Acta 1321: 243–251.

Bradford MM. 1976. A rapid and sensitive method for thequantitation of microgram quantities of protein utilizing theprinciple of protein–dye binding. Anal. Biochem. 72: 248–254.

Chang CS, Kirk RG, Lee P. 1998. Transient increase in the alpha3-isoform of Na,K-ATPase in rat erythroblastic cells.Histochem. J. 30: 811–818.

Green LC, Wagner DA, Glogowski J, Skipper PL, Wishnok JS,Tannenbaum SR. 1982. Analysis of nitrate, nitrite, and[15N]nitrate in biological fluids. Anal. Biochem. 126: 131–138.

Groot H, Hegi U, Sies H. 1993. Loss of alpha-tocopherol uponexposure to nitric oxide or the sydnonimine SIN-1. FEBSLett. 315: 139–142.

Hu YK, Eisses JF, Kaplan JH. 2000. Expression of an active Na,K-ATPase with an alpha-subunit lacking all twenty-three nativecysteine residues. J. Biol. Chem. 275: 30734–30739.

Kardos SA, Sultatos LG. 2000. Interactions of the organ-ophosphates paraoxon and methyl paraoxon with mousebrain acetylcholinesterase. Toxicol. Sci. 58: 118–126.

Kimelberg HG, Papahadjopoulos D. 1974. Effects of phospholipidacyl chain fluidity, phase transitions, and cholesterol on (Na+/K+)-stimulated adenosine triphosphatase. J. Biol. Chem. 249:1071–1080.

Miller RP, Farley RA. 1990. Beta subunit of (Na+/K+)-ATPase con-tains three disulfide bonds. Biochemistry 29: 1524–1532.

Muriel P, Mourelle M. 1990. The role of membrane composition

in ATPase activities of cirrhotic rats. Effect of silymarin. J.Appl. Toxicol. 10: 281–284.

Muriel P, Sandoval G. 2000a. Nitric oxide and peroxynitrite anionmodulate liver plasma membrane fluidity and Na+/K+-ATPaseactivity. Nitric Oxide: Biol. Chem. 4: 333–342.

Muriel P, Sandoval G. 2000b. Hepatic basolateral plasma high-affinity Ca2+-ATPase is inhibited by nitric oxide andperoxynitrite anion. J. Appl. Toxicol. 20: 435–342.

Noël F, Godfraind T. 1984. Heterogeneity of ouabain specificbinding sites and (Na+/K+)-ATPase inhibition in microsomesfrom rat heart. Biochem. Pharmacol. 33: 47–53.

Post RL, Merrit CR, Kinsolving CR, Albright CD. 1960. Membraneadenosine triphosphate as a participant in the active trans-port of sodium and potassium in the human erythrocyte. J.Biochem. Chem. 235: 1796–1802.

Radi R, Beckman JS, Bush KM, Freeman BA. 1991. Peroxynitriteoxidation of sulphydryls. The cytotoxic potential of super-oxide and nitric oxide. J. Biol. Chem. 266: 4244–4250.

Robinson JD. 1976. Substrate sites for the (Na+/K+)-dependentATPase. Biochim. Biophys. Acta 429: 1006–1019.

Rose HG, Oklander M. 1965. Improved procedure for the extrac-tion of lipids from human erythrocytes. J. Lipid Res. 6: 428–431.

Sato T, Kamata Y, Irifune M, Nishikawa T. 1997. Inhibitory effectof several nitric oxide-generating compounds on purifiedNa+/K+-ATPase activity from porcine cerebral cortex. J.Neurochem. 68: 1312–1318.

Shaffer JF, Han B, Chern WH, Lee FW. 1992. Lack of tolerance toa 24-hour infusion of s-nitroso n-acetylpenicillamine (SNAP)in conscious rabbits. J. Pharmacol. Exp. Ther. 260: 286–293.

Xie Z, Jack-Hays M, Wang Y, Periyasamy SM, Blanco G, HuangWH, Askari A. 1995. Different oxidant sensitivities of thealpha 1 and alpha 2 isoforms of Na+/K+-ATPase expressed inbaculovirus-infected insect cells. Biochem. Biophys. Res. Co.207: 155–159.