- Email: [email protected]
Dose-dependent reversal of digoxin-inhibited activity of an in vitro Na+K+ATPase model by digoxin-specific antibody
Toxicology Letters Toxleology Letters 85 (1996) 107-I 11
Dose-dependent reversal of digoxin-inhibited activity of an in vitro Na+K+ATPase model by digoxin-specific antibody Nathalie
J. Canoa, Alain E. Sabourauda, Marcel Debrayb, Jean-Michel G. Scherrmann* a
“INSERM U26 HGpital Femand W&d. 200 rue du Faubourg St-Denis, 75010 Paris, France bLaboratoire de Biostatistiques, FacultCde Phannacie. Paris, France
Received 22 May 1995;revised 15 January 1996;accepted 16 January 1996
We investigated the potency of digoxin-specific Fab fragments to reverse digoxin-induced Na+K+ATPase inhibition in rat brain microsomes according to (a) the extent of initial inhibition of Na+K+ATPaseand (b) the neutralizing
dose of antibody. Mathematical analysis of the digoxin concentration-Na+K+ATPase inhibition curve supports the existence of 2 digoxin sensitive Na+K+ATPase isofonns. The IC% was 1.3 x lOA M and 2.5 x 10m8M for the low (~1) and high (ar2) digoxin affinity isoenxyme, respectively. The reversal of digoxin-induced Na+K+ATPase inhibition was dependent on the ~digoxin-specific Fab concentration. The maximal effect was observed when the Fab:digoxin ratio was stoichiometrical and addition of an excess of antibodies did not result in a complete reversal of inhibition at the 4 digoxin concentrations studied. This simple and rapid in vitro model will be a useful tool to predict the efficacy of a new generation of antibodies. Keyworcis: Na+K+ATPase activity; Rat brain; Isoforms; Digoxin; Fab fragments
1. Introduction
Antibodies have been shown to inhibit the pharmacological or toxic effects of low-molecularweight compounds [l-4]. The most thoroughly studied antibodies that can reverse toxic effects are those that act against cardiac glycosides whose Abbreviations: Na+:K+ATPaae, sodium, potassium adenosine trlphosphataae; IC,, median inhibitory concentration; EC%, median effect concentration. l Correspending autha’r,Tel: 140 05 43 46; Fax: 40 34 40 64.
pharmacological effects are due to the inhibition of membrane-bound Na+K+ATPase [5-71. The restoration of enzyme activity is related to the dissociation of digoxin from Na+K+ATPase receptor binding sites [8]. In vivo models requiring complex experiments have described the pharmacological recovery following administration of antibodies against cardiac glycosides [5,9]. However, no simple in vitro models are currently available to predict the potency of such antibodies, or new generation antibodies, for restoring receptor activity. With this aim, we investigated the potency
0378-4274/96/$15.000 1996Elsevier Science Ireland Ltd. All rights reserved PII 0378-4274(96)03647-S
108
N.J.Cano et al. /Toxicology
of digoxin-specific Fab fragments to reverse digoxin-induced Na+K+ATPase inhibition in rat brain microsomes according to (a) the extent of initial inhibition of Na+K+ATPase activity by using 4 different digoxin concentrations and (b) the neutralizing concentration of antibody. 2. Materials and methds 2.1. Reagents and animals Male Shermann rats (60 days old) were purchased from Janvier (Le Genest, France). All chemical products and enzymes were obtained from Sigma (L’Isle d’Abeau Chesnes, France). Polyclonal digoxin-specific Fab fragments (Digidote) were purchased from Boehringer GmbH (Mannheim, Germany). 2.2. Preparation of microsomes The brain membrane fraction was prepared by differential centrifugation using the method of Sweadner [lo] as modified by Berrebi-Bertrand et al. [l 11.Briefly, rat brains were homogenized with 5 strokes of a Teflon-glass homogenizer in 15 ml 0.32 M sucrose containing 1 mM EDTA, 0.1 FM phenylmethylsulfonyl fluoride and 30 mM imidazole-HCl, pH 7.2. The homogenate was centrifuged at 850 x g for 20 min (Sorvall SS34 rotor) to remove nuclei and unhomogenized material, then at 8500 x g (Sorvall SS34 rotor) for 20 min to remove mitochondria and myelin fragments. The supematant was centrifuged at 100 000 x g (Kontron TFT 65-38 rotor) for 30 min. The pellet, fraction enriched in i.e. the microsomal Na+Ka+ATPase activity, was resuspended in the buffer, 0.32 M sucrose, 1 mM EDTA, 30 mM imidazole-HCl pH 7.2, to obtain a protein concentration of about 2.5 mg/ml, aliquoted and stored at -70°C until use. All microsomal samples were used within 3 weeks of preparation. Protein concentrations were determined according to Lowry et al. [12] with serum albumin as standard.
2.3. Sensitivity of Na+K+ATPase to digoxin Na+K+ATPase activity was determined using
Letters 85 (19%)
107-111
the method described by Skou [13] modified by Lelibvre et al. [14]. The activity was measured in an ATP regenerating medium by continuously recording NADH oxidation at 340 nm, using a Uvikon 930 spectrophotometer (Kontron, Montigny le Bretonneux, France). Enzyme activity was linearly measured as a function of the amount of microsomal proteins from 0.1 to 3 pg. The enzymatic assay was carried out at 37°C with native microsomes without detergent treatment. Each cell contained (final volume 0.6 ml) 100 mM NaCI, 2 mM phosphoenolpyruvate, 10 mM KCl, 4 mM ATP, 4 mM MgC12, 30 mM imidazole-HCl (PH 7.2), 0.4 mM NADH, 3.5 units pyruvate kinase and 5 units lactate dehydrogenase. Na+K+ATPase activity was studied at digoxin concentrations ranging from 2.1 x 10V4 M to 2.1 x 10m9M. The digoxin solution (1.28 x lo-* M) was dissolved in ethanol, further dilutions were made in mixed solution of water:ethanol with progressively decreasing percentage of ethanol. The enzymatic reaction was started by addition of microsomes and was continued for a 26-min period, with continuous stirring. The non-specific binding was determined with a 3 x 10m4M digoxigenin solution which was more lypophilic than digoxin. The specific Na+K+ATPase activity was about 60 pmol/Pi/ h/mg in brain microsomes. The orientation of the microsome was determined in presence of digoxin and ouabain, 82% were leaky and permeable to both molecules and 18% were inside out and were impermeable to ouabain. The right inside out pop ulation activity was not observed because this population was impermeable to ATP. 2.4. Effect of digoxin-specific Fab fragments on Na+K+ATPase activity
The reversal of digoxin-induced Na+K+ATPase inhibition by addition of digoxin-specific Fab fragments was studied at digoxin concentrations of 7 x lo-*, 2.1 x 10-7, 7 x 10-7, 2.1 x lO-‘j M. At each digoxin concentration, digoxin-specific Fab fragments were added to the incubation medium, 2 min after the start of the reaction, at a Fabldigoxin stoichiometrical ratio ranging from 1:8 to 2: 1. Preliminary experiments have
N.J. Cano et al. /Toxicology Letters 85 (19%) 107-111
demonstrated that addition of non-specific Fab fragments at a concentration equal to that of digoxin-specific Fab fragments had no incidence on the total activity or on the digoxin-induced inhibition of Na+K+ATPase activity. 2.5. Data analysis For each digoxin concentration, inhibition was calculated as a percentage of the specific Na+K+ATPase activity. For the relationship between Na+K+ATPase activity and digoxin concentration, the curve fitted to the experimental data was obtained using the following equation as described by BerrebiBertrand et al. [II]: i=n A = C (A~(IC5di)I((1C di + C) i=l where A is the alctivity, n is the number of Na+K+ATPase population, Ai is the maximum activity of the i population with Ai = ClOO, IC, is the concentration for half-maximum inhibition of the i population and C is the digoxin concentration. The number of independent sites to lit the data was determined according to the Schwarz criterion. For the relationship between Na+K+ATPase activity and Fab concentration, the curve was fitted according to the following equation: A = A, + ((A,,
- A,).CB)/(EC5/
+ CB)
where A is the activity, A, is the activity without Fab fragments, A,,, is the maximum activity with Fab fragment, C is the Fab fragment concentration, /3 is the Hill #coefficient for digoxin binding and EC, is the Fab fragment concentration for half-maximum activity restoration. All the parameters were estimated by non-linear regression using a MKModel Software (Biosoft, Cambridge, USA). 3. Results and discussion Digitalis-specific antibodies have been demon-
109
strated to reverse the digitalis-induced Na+K+ATPase inhibition. Restoration of human erythrocyte potassium influx [ 151 and ATP cleavage in canine myocardial microsomes [16] supports the restoration of Na+K+ATPase enzyme activity by digoxin-specific antibodies. A direct interaction of antibodies with membranebound digitalis is unlikely because such detoxication processes have been described for intracellular toxins such as colchicine [ 171and trichothecene mycotoxin Tz [18]. In fact, specific digitalis antibodies lower the free extracellular cardiac glycoside concentration and create a net eMux from cells by altering the equilibrium between cellular and extracellular digitalis. To our knowledge, the dose-effect relationship of digoxin-specific Fab fragments has not been investigated in vitro or in vivo. In this study, we investigated the inhibition of rat brain microsome Na+K+ATPase activity by digoxin and its restoration by addition of digoxin-specific Fab fragments. Despite the fact that the rat is relatively insensitive to the cardiac effect of digitalis because most Na+K+ATPase molecules outside the central nervous system contain the crl isoform which has a low affinity for digoxin, rat brain microsomes were chosen for the development of the in vitro model because the dissociation constants of rat brain ATPase for digitalis (Kdl = 1.8 x 10m7 M and &z = 3.4 x 10s8 M) (data not shown) was similar to those observed in dog heart (Kd = 10T7 M) [19], human heart (&r = 1.7 x 10m7M and & = 2 x 10s8 M) and beef heart (& = 1.7 x 1O-7 M) [20]. First, Na+K+ATPase activity of rat brain microsomes was studied as a function of digoxin concentrations (Fig. 1). The dose/response curve spanned 6 orders of magnitude and indicated the presence of more than one class of enzyme binding site. The best lit for the curve was obtained assuming the existence of 2 inhibitory processes. The I& was 1.3 x 10m4M and 2.5 x lo-* M for the low (arl) and high (ar2) digoxin affinity isoenzyme, respectively. The low and high affinity isoenzymes accounted, respectively, for 3 1.7 and 68.3% of the specific Na+K+ATPase activity. Heterogeneity of ouabain sites has been reported in rat brain Na+K+ATPase and related to 2 distinct isoen-
N.J. Cam et al. /Toxicology Letters 85 (19%) 107-111
110
~.. .
to 0
lo-9
--..__ __
1o-7
1c*
10-6 1o-5
1o-4
1o-3
DigoxinconcentrationImol/L)
Fig. 1. Digoxin inhibition of rat brain membrane Na+K+ATPase activity. Values are mean t S.D. of 10 experiments. Each experiment was done in duplicate. The data were analysed by the non-linear regression model (Materials and methods). The solid line represents the theoretical curve assuming a two-site model tit, the theoretical curve for each site was plotted as follows: ---, for high digoxin affinity; - - - - -, for low digoxin at?inity.
zymes [IO]. The presence of 3 isoenzymes, (~1, ~2, ar3 encoded by separate genes, has been detected by immunological experiments [21-231. In our rat brain enzyme preparation, the activity of ouabainsensitive (~2 and ar3 isoforms could not be distinguished. In the present study, statistical analysis supports the existence of only 2 digoxinsensitive Na+K+ATPase isoforms.
Second, the influence of digoxin-specific antibody Fab fragments on digoxin-induced Na+K+ATPase inhibition was studied only for digoxin concentration ranging from 7 x 10e8 to 2.1 x 10S6 M because higher concentrations would have required too much Fab. Over this concentration range, only the activity of the high aflinity isoenzyme was inhibited, the low affmity Na+K+ATPase activity remained maximal (Fig. reversal of 1). The digoxin-induced Na+K+ATPase inhibition was dependent on the digoxin-specific Fab concentration (Fig. 2). Except at the lower digoxin concentration (7 x 10-s M), the addition of excess antibodies (Fab:digoxin 2: 1) did not result in a complete reversal of inhibition. Whatever the digoxin concentration, the maxima1 effect was observed when the Fab:digoxin ratio was stoichiometrical. Recovery of the high affinity isoenzyme activity was 100,83,70 and 50% while initial activity was 30, 11, 3 and O%, respectively. Thus, recovery of Na+K+ATPase activity depends on the extent of its initial inhibition. There was no statistical difference between EC% values (0.19 f 0.11, 0.32 * 0.015, 0.41 f 0.03 and 0.39 * 0.02 M) at the 4 digoxin concentrations studied (7 x 10m8, 2.1 x 10S7, 7 x 10m7, 2.1 x 10m6 M, respectively). The inability of digoxin-specific antibodies to completely reverse digoxin-induced Na+K+ATPase inhibition has also been reported for a beef heart microsome preparation treated with 2.4 x 10S7 M digoxin
PI. The incomplete reversal of digoxin-induced
2ot
IIt_ 0.0
I
0.5
1.0 Fab
: digoxh
1.5
2.0
:: q
] ; 2.5
ratio
Fig. 2. Reversal of digoxin-induced Na+K+ATPase inhibition after addition of digoxin-specitic Fab fragments was studied at 4 digoxin concentrations, (0) 2.1 x 10m6M, (0) 7 x IO-’ M, ( m 2.1 x IO-’ M, (Cl) 7 x IO-* M. Values are mean f SD. of LO-30 experiments. Each experiment was done in duplicate.
Na+K+ATPase inhibition does not depend on the affinities of the receptor-digoxin complex vs. the Fab-digoxin complex because Fab fragment affinity was 250-fold higher than Na+K+ATPase affinity for digoxin (data not shown). On the other hand, the heterogeneity of the microsome population involves the digoxin binding to the external and internal face of the membrane, so the intracellular binding of digoxin could limit the exchange of digoxin between the inside and outside compartments and the accessibility of Fab fragments to digoxin. Moreover, as the experimental time corresponding to the enzyme turnover (26 min) was too short compared with the dissociation halflife of digoxin from Na+K+ATPase (40 min) (data
NJ.
Cano et ai. /Toxicology
not shown), the equilibrium between the Fab fragments-digoxin vs. Na+K+ATPase-digoxin complexes was not achieved, explaining the incomplete reversal of Na+K+ATPase activity. Nevertheless, this in vitro Na+K+ATPase model permits the investigation of the potency of digoxin-specific antibodies for the restoration of receptor activity. ‘When the new generation of fragments smaller than Fab, like the Fv fragment [24], become avaihble, a simple and rapid model such as the one described here will be useful before more complete in vivo evaluation. Acknowledgements This work was sulpported by grants from Institut National de la Sante et de la Recherche Medicale. The authors wish to thank Mr. A. Strickland for technical assistance in the preparation of the manuscript. References VI Singer, E. (1942) Serological protection against arsenic compounds. Aust. J. Exp. Biol. Med. Sci. 20, 209-212.
121Creech, H.J. (1952) Chemical and immunological properties of carcinogen-protein conjugates. Cancer Res. 12, 557-564. I31 Johnson, H.M., Fret, P.A., Angelotti, R., Campeh, J.E. and Lewis, K.A. (1964) Haptenic properties of paralytic shellfish poison conjugated to proteins by formaldehyde treatment. Proc. S’oc. Exp. Biol. Med. 117, 425-430. [41 Peck, R.M. and Peck, E.B. (1977) Inhibition of chemically induced neoplasia by immunization with an antigenic conjugate. Cancer Res. 31, carcinogen-protein 1550-1554. I51 Schmidt, D.H. and Butler, V.P. (1971) Immunological protection against digoxin toxicity. 1. Clin. Invest. 50, 866-871. PI Curd, J., Smith, T.W., Jaton, J.C. and Haber, E. (1971) The isolation of digoxin-specific antibody and its use in reversing the effects of digoxin. Proc. Nat]. Acad. Sci. USA 68,2401-2406. 171 Smith, T.W., Haber, E., Yeatman, L. and Butler, V.P. (1976) Reversal of advanced digoxin intoxication with Fab fragments of digoxin-specific antibodies. N. En@. J. Med. 294, 797-800. 181 Bossaller, C. (I 98 I) The effect of digitalis glycosidesspecific antisera on the binding of the glycosides to Na+K+ATPase. Arxneim.-Forsch. Drug Res. 31, 625-628. 191 Lechat, P., Mudgett-Hunter, M., Margohes, M.N., Haber, E. and Smith, T.W. (1984) Reversal of lethal
Letters 85 (19963 iO?-111
111
digoxintoxicity in guinea pigs using monoclonal antibodies and Fab fragments. J. Pharmacol. Exp. Ther. 229, 210-213. WI Sweadner, K.J. (1979) Two molecular forms of Na+K+ATPase in brain. J. Biol. Chem. 254,6060+X7. 1111Berrebi-Bertrand, I., Maixent, J.M., Christe, G. and Lelitre, L.G. (1990) Two active Na+K+ATPases of high affinity for ouabain in adult rat brain membranes. B&him. Biophys. Acta 1021, 148-156. 1121 Lowry, O.H., Rosebrough, N.J., Farr, A.L. and Randall, R.J. (1951) Protein measurement with the Folin phenol reagent. J. Biol. Chem. 193, 265-275. I131 Skou, J.C. (1960) The influence of some cations on an adenosine triphosphatase from peripheral nerves. Biochem. Biophys. Acta 23, 394-401. P41 Leli&.vre,L.G., Maixent, J.M., Lorente, P., Mouas, C., Charlemagne, D. and Swynghedauw, B. (1986) Prolonged responsiveness to ouabain in hypertrophied rat heart: physiological and biochemical evidence. Am. J. Physiol. 251, H923-H93 1. [I51 Gardner, J.D., Kiino, D.R., Swarm, T.J. and Butler, V.P. (1973) Effects of digoxin-specific antibodies on accumulation and binding of digoxin by human erythrocytes. J. Clin. Invest. 52, 1820-1833. U61 Smith, T.W. (1974) Use of antibodies in the study of the mechanism of action of digitalis. Ann. N.Y. Acad. Sci. 242, 731-736. 1171 Rouan, SK., Ottemess, I.G., Cunningham, A.C., Holden, H.E. and Rhodes, CT. (1990) Reversal of colchicine-induced ntitotic arrest in Chinese hamster cells with a colchicine specific monoclonal antibody. Am. J. Pathol. 137, 779-787. WI Hunter, K.W., Brimtield, A.A., Knower, A.T., Powell, J.A. and Feuerstein, G.Z. (1990) Reversal of intracellular toxicity of the trichothecene mycotoxin T-2 with monoclonal antibody. J. Pharmacol. Exp. Ther. 255, 1183-1187. 1191 Songu-mize, E., Gunter, I. and Caldwell, W. (1989) Comparative ability of digoxin and an aminosugar cardiac glycoside to bind to and inhibit Na+K+adenosine triphosphatase. B&hem. Pharmacol. 38, 3689-3695. [201 Erdmann, E., Werdan, K. and Brown, L. (1985) Multiplicity of cardiac glycoside receptors in the heart. Trends B&hem. Sci. July, 293-295. 1211 Shull, G.E., Greeb, J. and Lingrel, J.B. (1986) Molecular cloning of three distinct forms of the Na+K+ATPase OLsubunit from rat brain. Biochemistry 25, 8125-8132. 1221 Young, R.M. and Lingrel, J.B. (1987) Tissue distribution of mRNAs encoding the Q isoforms and fl subunit of rat Na+K+ATPase. Biochem. Biophys. Res. Commun. 145: 52-58. [231 Urayama, 0. and Sweadner, K.J. (1988) Ouabain sensitivity of the c+ isoxyme of rat Na,K-ATPase. B&hem. Biophys. Res. Commun. 156, 796-800. [241 Anthony, J., Near, R., Wong, S.L., Iida, E., Ernst, E., Wittekind, M. and Haber, E. (1992) Production of stable antidigoxin Fv in Escherichia cob. Mol. Immunol. 29, 1237-1247.
Dose-dependent reversal of digoxin-inhibited activity of an in vitro Na+K+ATPase model by digoxin-specific antibody Nathalie
J. Canoa, Alain E. Sabourauda, Marcel Debrayb, Jean-Michel G. Scherrmann* a
“INSERM U26 HGpital Femand W&d. 200 rue du Faubourg St-Denis, 75010 Paris, France bLaboratoire de Biostatistiques, FacultCde Phannacie. Paris, France
Received 22 May 1995;revised 15 January 1996;accepted 16 January 1996
We investigated the potency of digoxin-specific Fab fragments to reverse digoxin-induced Na+K+ATPase inhibition in rat brain microsomes according to (a) the extent of initial inhibition of Na+K+ATPaseand (b) the neutralizing
dose of antibody. Mathematical analysis of the digoxin concentration-Na+K+ATPase inhibition curve supports the existence of 2 digoxin sensitive Na+K+ATPase isofonns. The IC% was 1.3 x lOA M and 2.5 x 10m8M for the low (~1) and high (ar2) digoxin affinity isoenxyme, respectively. The reversal of digoxin-induced Na+K+ATPase inhibition was dependent on the ~digoxin-specific Fab concentration. The maximal effect was observed when the Fab:digoxin ratio was stoichiometrical and addition of an excess of antibodies did not result in a complete reversal of inhibition at the 4 digoxin concentrations studied. This simple and rapid in vitro model will be a useful tool to predict the efficacy of a new generation of antibodies. Keyworcis: Na+K+ATPase activity; Rat brain; Isoforms; Digoxin; Fab fragments
1. Introduction
Antibodies have been shown to inhibit the pharmacological or toxic effects of low-molecularweight compounds [l-4]. The most thoroughly studied antibodies that can reverse toxic effects are those that act against cardiac glycosides whose Abbreviations: Na+:K+ATPaae, sodium, potassium adenosine trlphosphataae; IC,, median inhibitory concentration; EC%, median effect concentration. l Correspending autha’r,Tel: 140 05 43 46; Fax: 40 34 40 64.
pharmacological effects are due to the inhibition of membrane-bound Na+K+ATPase [5-71. The restoration of enzyme activity is related to the dissociation of digoxin from Na+K+ATPase receptor binding sites [8]. In vivo models requiring complex experiments have described the pharmacological recovery following administration of antibodies against cardiac glycosides [5,9]. However, no simple in vitro models are currently available to predict the potency of such antibodies, or new generation antibodies, for restoring receptor activity. With this aim, we investigated the potency
0378-4274/96/$15.000 1996Elsevier Science Ireland Ltd. All rights reserved PII 0378-4274(96)03647-S
108
N.J.Cano et al. /Toxicology
of digoxin-specific Fab fragments to reverse digoxin-induced Na+K+ATPase inhibition in rat brain microsomes according to (a) the extent of initial inhibition of Na+K+ATPase activity by using 4 different digoxin concentrations and (b) the neutralizing concentration of antibody. 2. Materials and methds 2.1. Reagents and animals Male Shermann rats (60 days old) were purchased from Janvier (Le Genest, France). All chemical products and enzymes were obtained from Sigma (L’Isle d’Abeau Chesnes, France). Polyclonal digoxin-specific Fab fragments (Digidote) were purchased from Boehringer GmbH (Mannheim, Germany). 2.2. Preparation of microsomes The brain membrane fraction was prepared by differential centrifugation using the method of Sweadner [lo] as modified by Berrebi-Bertrand et al. [l 11.Briefly, rat brains were homogenized with 5 strokes of a Teflon-glass homogenizer in 15 ml 0.32 M sucrose containing 1 mM EDTA, 0.1 FM phenylmethylsulfonyl fluoride and 30 mM imidazole-HCl, pH 7.2. The homogenate was centrifuged at 850 x g for 20 min (Sorvall SS34 rotor) to remove nuclei and unhomogenized material, then at 8500 x g (Sorvall SS34 rotor) for 20 min to remove mitochondria and myelin fragments. The supematant was centrifuged at 100 000 x g (Kontron TFT 65-38 rotor) for 30 min. The pellet, fraction enriched in i.e. the microsomal Na+Ka+ATPase activity, was resuspended in the buffer, 0.32 M sucrose, 1 mM EDTA, 30 mM imidazole-HCl pH 7.2, to obtain a protein concentration of about 2.5 mg/ml, aliquoted and stored at -70°C until use. All microsomal samples were used within 3 weeks of preparation. Protein concentrations were determined according to Lowry et al. [12] with serum albumin as standard.
2.3. Sensitivity of Na+K+ATPase to digoxin Na+K+ATPase activity was determined using
Letters 85 (19%)
107-111
the method described by Skou [13] modified by Lelibvre et al. [14]. The activity was measured in an ATP regenerating medium by continuously recording NADH oxidation at 340 nm, using a Uvikon 930 spectrophotometer (Kontron, Montigny le Bretonneux, France). Enzyme activity was linearly measured as a function of the amount of microsomal proteins from 0.1 to 3 pg. The enzymatic assay was carried out at 37°C with native microsomes without detergent treatment. Each cell contained (final volume 0.6 ml) 100 mM NaCI, 2 mM phosphoenolpyruvate, 10 mM KCl, 4 mM ATP, 4 mM MgC12, 30 mM imidazole-HCl (PH 7.2), 0.4 mM NADH, 3.5 units pyruvate kinase and 5 units lactate dehydrogenase. Na+K+ATPase activity was studied at digoxin concentrations ranging from 2.1 x 10V4 M to 2.1 x 10m9M. The digoxin solution (1.28 x lo-* M) was dissolved in ethanol, further dilutions were made in mixed solution of water:ethanol with progressively decreasing percentage of ethanol. The enzymatic reaction was started by addition of microsomes and was continued for a 26-min period, with continuous stirring. The non-specific binding was determined with a 3 x 10m4M digoxigenin solution which was more lypophilic than digoxin. The specific Na+K+ATPase activity was about 60 pmol/Pi/ h/mg in brain microsomes. The orientation of the microsome was determined in presence of digoxin and ouabain, 82% were leaky and permeable to both molecules and 18% were inside out and were impermeable to ouabain. The right inside out pop ulation activity was not observed because this population was impermeable to ATP. 2.4. Effect of digoxin-specific Fab fragments on Na+K+ATPase activity
The reversal of digoxin-induced Na+K+ATPase inhibition by addition of digoxin-specific Fab fragments was studied at digoxin concentrations of 7 x lo-*, 2.1 x 10-7, 7 x 10-7, 2.1 x lO-‘j M. At each digoxin concentration, digoxin-specific Fab fragments were added to the incubation medium, 2 min after the start of the reaction, at a Fabldigoxin stoichiometrical ratio ranging from 1:8 to 2: 1. Preliminary experiments have
N.J. Cano et al. /Toxicology Letters 85 (19%) 107-111
demonstrated that addition of non-specific Fab fragments at a concentration equal to that of digoxin-specific Fab fragments had no incidence on the total activity or on the digoxin-induced inhibition of Na+K+ATPase activity. 2.5. Data analysis For each digoxin concentration, inhibition was calculated as a percentage of the specific Na+K+ATPase activity. For the relationship between Na+K+ATPase activity and digoxin concentration, the curve fitted to the experimental data was obtained using the following equation as described by BerrebiBertrand et al. [II]: i=n A = C (A~(IC5di)I((1C di + C) i=l where A is the alctivity, n is the number of Na+K+ATPase population, Ai is the maximum activity of the i population with Ai = ClOO, IC, is the concentration for half-maximum inhibition of the i population and C is the digoxin concentration. The number of independent sites to lit the data was determined according to the Schwarz criterion. For the relationship between Na+K+ATPase activity and Fab concentration, the curve was fitted according to the following equation: A = A, + ((A,,
- A,).CB)/(EC5/
+ CB)
where A is the activity, A, is the activity without Fab fragments, A,,, is the maximum activity with Fab fragment, C is the Fab fragment concentration, /3 is the Hill #coefficient for digoxin binding and EC, is the Fab fragment concentration for half-maximum activity restoration. All the parameters were estimated by non-linear regression using a MKModel Software (Biosoft, Cambridge, USA). 3. Results and discussion Digitalis-specific antibodies have been demon-
109
strated to reverse the digitalis-induced Na+K+ATPase inhibition. Restoration of human erythrocyte potassium influx [ 151 and ATP cleavage in canine myocardial microsomes [16] supports the restoration of Na+K+ATPase enzyme activity by digoxin-specific antibodies. A direct interaction of antibodies with membranebound digitalis is unlikely because such detoxication processes have been described for intracellular toxins such as colchicine [ 171and trichothecene mycotoxin Tz [18]. In fact, specific digitalis antibodies lower the free extracellular cardiac glycoside concentration and create a net eMux from cells by altering the equilibrium between cellular and extracellular digitalis. To our knowledge, the dose-effect relationship of digoxin-specific Fab fragments has not been investigated in vitro or in vivo. In this study, we investigated the inhibition of rat brain microsome Na+K+ATPase activity by digoxin and its restoration by addition of digoxin-specific Fab fragments. Despite the fact that the rat is relatively insensitive to the cardiac effect of digitalis because most Na+K+ATPase molecules outside the central nervous system contain the crl isoform which has a low affinity for digoxin, rat brain microsomes were chosen for the development of the in vitro model because the dissociation constants of rat brain ATPase for digitalis (Kdl = 1.8 x 10m7 M and &z = 3.4 x 10s8 M) (data not shown) was similar to those observed in dog heart (Kd = 10T7 M) [19], human heart (&r = 1.7 x 10m7M and & = 2 x 10s8 M) and beef heart (& = 1.7 x 1O-7 M) [20]. First, Na+K+ATPase activity of rat brain microsomes was studied as a function of digoxin concentrations (Fig. 1). The dose/response curve spanned 6 orders of magnitude and indicated the presence of more than one class of enzyme binding site. The best lit for the curve was obtained assuming the existence of 2 inhibitory processes. The I& was 1.3 x 10m4M and 2.5 x lo-* M for the low (arl) and high (ar2) digoxin affinity isoenzyme, respectively. The low and high affinity isoenzymes accounted, respectively, for 3 1.7 and 68.3% of the specific Na+K+ATPase activity. Heterogeneity of ouabain sites has been reported in rat brain Na+K+ATPase and related to 2 distinct isoen-
N.J. Cam et al. /Toxicology Letters 85 (19%) 107-111
110
~.. .
to 0
lo-9
--..__ __
1o-7
1c*
10-6 1o-5
1o-4
1o-3
DigoxinconcentrationImol/L)
Fig. 1. Digoxin inhibition of rat brain membrane Na+K+ATPase activity. Values are mean t S.D. of 10 experiments. Each experiment was done in duplicate. The data were analysed by the non-linear regression model (Materials and methods). The solid line represents the theoretical curve assuming a two-site model tit, the theoretical curve for each site was plotted as follows: ---, for high digoxin affinity; - - - - -, for low digoxin at?inity.
zymes [IO]. The presence of 3 isoenzymes, (~1, ~2, ar3 encoded by separate genes, has been detected by immunological experiments [21-231. In our rat brain enzyme preparation, the activity of ouabainsensitive (~2 and ar3 isoforms could not be distinguished. In the present study, statistical analysis supports the existence of only 2 digoxinsensitive Na+K+ATPase isoforms.
Second, the influence of digoxin-specific antibody Fab fragments on digoxin-induced Na+K+ATPase inhibition was studied only for digoxin concentration ranging from 7 x 10e8 to 2.1 x 10S6 M because higher concentrations would have required too much Fab. Over this concentration range, only the activity of the high aflinity isoenzyme was inhibited, the low affmity Na+K+ATPase activity remained maximal (Fig. reversal of 1). The digoxin-induced Na+K+ATPase inhibition was dependent on the digoxin-specific Fab concentration (Fig. 2). Except at the lower digoxin concentration (7 x 10-s M), the addition of excess antibodies (Fab:digoxin 2: 1) did not result in a complete reversal of inhibition. Whatever the digoxin concentration, the maxima1 effect was observed when the Fab:digoxin ratio was stoichiometrical. Recovery of the high affinity isoenzyme activity was 100,83,70 and 50% while initial activity was 30, 11, 3 and O%, respectively. Thus, recovery of Na+K+ATPase activity depends on the extent of its initial inhibition. There was no statistical difference between EC% values (0.19 f 0.11, 0.32 * 0.015, 0.41 f 0.03 and 0.39 * 0.02 M) at the 4 digoxin concentrations studied (7 x 10m8, 2.1 x 10S7, 7 x 10m7, 2.1 x 10m6 M, respectively). The inability of digoxin-specific antibodies to completely reverse digoxin-induced Na+K+ATPase inhibition has also been reported for a beef heart microsome preparation treated with 2.4 x 10S7 M digoxin
PI. The incomplete reversal of digoxin-induced
2ot
IIt_ 0.0
I
0.5
1.0 Fab
: digoxh
1.5
2.0
:: q
] ; 2.5
ratio
Fig. 2. Reversal of digoxin-induced Na+K+ATPase inhibition after addition of digoxin-specitic Fab fragments was studied at 4 digoxin concentrations, (0) 2.1 x 10m6M, (0) 7 x IO-’ M, ( m 2.1 x IO-’ M, (Cl) 7 x IO-* M. Values are mean f SD. of LO-30 experiments. Each experiment was done in duplicate.
Na+K+ATPase inhibition does not depend on the affinities of the receptor-digoxin complex vs. the Fab-digoxin complex because Fab fragment affinity was 250-fold higher than Na+K+ATPase affinity for digoxin (data not shown). On the other hand, the heterogeneity of the microsome population involves the digoxin binding to the external and internal face of the membrane, so the intracellular binding of digoxin could limit the exchange of digoxin between the inside and outside compartments and the accessibility of Fab fragments to digoxin. Moreover, as the experimental time corresponding to the enzyme turnover (26 min) was too short compared with the dissociation halflife of digoxin from Na+K+ATPase (40 min) (data
NJ.
Cano et ai. /Toxicology
not shown), the equilibrium between the Fab fragments-digoxin vs. Na+K+ATPase-digoxin complexes was not achieved, explaining the incomplete reversal of Na+K+ATPase activity. Nevertheless, this in vitro Na+K+ATPase model permits the investigation of the potency of digoxin-specific antibodies for the restoration of receptor activity. ‘When the new generation of fragments smaller than Fab, like the Fv fragment [24], become avaihble, a simple and rapid model such as the one described here will be useful before more complete in vivo evaluation. Acknowledgements This work was sulpported by grants from Institut National de la Sante et de la Recherche Medicale. The authors wish to thank Mr. A. Strickland for technical assistance in the preparation of the manuscript. References VI Singer, E. (1942) Serological protection against arsenic compounds. Aust. J. Exp. Biol. Med. Sci. 20, 209-212.
121Creech, H.J. (1952) Chemical and immunological properties of carcinogen-protein conjugates. Cancer Res. 12, 557-564. I31 Johnson, H.M., Fret, P.A., Angelotti, R., Campeh, J.E. and Lewis, K.A. (1964) Haptenic properties of paralytic shellfish poison conjugated to proteins by formaldehyde treatment. Proc. S’oc. Exp. Biol. Med. 117, 425-430. [41 Peck, R.M. and Peck, E.B. (1977) Inhibition of chemically induced neoplasia by immunization with an antigenic conjugate. Cancer Res. 31, carcinogen-protein 1550-1554. I51 Schmidt, D.H. and Butler, V.P. (1971) Immunological protection against digoxin toxicity. 1. Clin. Invest. 50, 866-871. PI Curd, J., Smith, T.W., Jaton, J.C. and Haber, E. (1971) The isolation of digoxin-specific antibody and its use in reversing the effects of digoxin. Proc. Nat]. Acad. Sci. USA 68,2401-2406. 171 Smith, T.W., Haber, E., Yeatman, L. and Butler, V.P. (1976) Reversal of advanced digoxin intoxication with Fab fragments of digoxin-specific antibodies. N. En@. J. Med. 294, 797-800. 181 Bossaller, C. (I 98 I) The effect of digitalis glycosidesspecific antisera on the binding of the glycosides to Na+K+ATPase. Arxneim.-Forsch. Drug Res. 31, 625-628. 191 Lechat, P., Mudgett-Hunter, M., Margohes, M.N., Haber, E. and Smith, T.W. (1984) Reversal of lethal
Letters 85 (19963 iO?-111
111
digoxintoxicity in guinea pigs using monoclonal antibodies and Fab fragments. J. Pharmacol. Exp. Ther. 229, 210-213. WI Sweadner, K.J. (1979) Two molecular forms of Na+K+ATPase in brain. J. Biol. Chem. 254,6060+X7. 1111Berrebi-Bertrand, I., Maixent, J.M., Christe, G. and Lelitre, L.G. (1990) Two active Na+K+ATPases of high affinity for ouabain in adult rat brain membranes. B&him. Biophys. Acta 1021, 148-156. 1121 Lowry, O.H., Rosebrough, N.J., Farr, A.L. and Randall, R.J. (1951) Protein measurement with the Folin phenol reagent. J. Biol. Chem. 193, 265-275. I131 Skou, J.C. (1960) The influence of some cations on an adenosine triphosphatase from peripheral nerves. Biochem. Biophys. Acta 23, 394-401. P41 Leli&.vre,L.G., Maixent, J.M., Lorente, P., Mouas, C., Charlemagne, D. and Swynghedauw, B. (1986) Prolonged responsiveness to ouabain in hypertrophied rat heart: physiological and biochemical evidence. Am. J. Physiol. 251, H923-H93 1. [I51 Gardner, J.D., Kiino, D.R., Swarm, T.J. and Butler, V.P. (1973) Effects of digoxin-specific antibodies on accumulation and binding of digoxin by human erythrocytes. J. Clin. Invest. 52, 1820-1833. U61 Smith, T.W. (1974) Use of antibodies in the study of the mechanism of action of digitalis. Ann. N.Y. Acad. Sci. 242, 731-736. 1171 Rouan, SK., Ottemess, I.G., Cunningham, A.C., Holden, H.E. and Rhodes, CT. (1990) Reversal of colchicine-induced ntitotic arrest in Chinese hamster cells with a colchicine specific monoclonal antibody. Am. J. Pathol. 137, 779-787. WI Hunter, K.W., Brimtield, A.A., Knower, A.T., Powell, J.A. and Feuerstein, G.Z. (1990) Reversal of intracellular toxicity of the trichothecene mycotoxin T-2 with monoclonal antibody. J. Pharmacol. Exp. Ther. 255, 1183-1187. 1191 Songu-mize, E., Gunter, I. and Caldwell, W. (1989) Comparative ability of digoxin and an aminosugar cardiac glycoside to bind to and inhibit Na+K+adenosine triphosphatase. B&hem. Pharmacol. 38, 3689-3695. [201 Erdmann, E., Werdan, K. and Brown, L. (1985) Multiplicity of cardiac glycoside receptors in the heart. Trends B&hem. Sci. July, 293-295. 1211 Shull, G.E., Greeb, J. and Lingrel, J.B. (1986) Molecular cloning of three distinct forms of the Na+K+ATPase OLsubunit from rat brain. Biochemistry 25, 8125-8132. 1221 Young, R.M. and Lingrel, J.B. (1987) Tissue distribution of mRNAs encoding the Q isoforms and fl subunit of rat Na+K+ATPase. Biochem. Biophys. Res. Commun. 145: 52-58. [231 Urayama, 0. and Sweadner, K.J. (1988) Ouabain sensitivity of the c+ isoxyme of rat Na,K-ATPase. B&hem. Biophys. Res. Commun. 156, 796-800. [241 Anthony, J., Near, R., Wong, S.L., Iida, E., Ernst, E., Wittekind, M. and Haber, E. (1992) Production of stable antidigoxin Fv in Escherichia cob. Mol. Immunol. 29, 1237-1247.