Effects of (Na+ + K+)-ATPase-specific antibodies of enzymatic activity and monovalent cation transport

Epitope mapping by amino-acid-sequence-specific antibodies reveals that both ends of the alpha subunit of Na+/K(+)-ATPase are located on the cytoplasmic side of the membrane

Right-side-out vesicles of pig kidney microsomes and amino-acid-sequence-specific antibodies were used to probe the sidedness of the C-terminus and the N-terminus of the catalytic alpha subunit of Na+/K(+)-ATPase. Polyclonal antibodies were raised in rabbits against the peptide corresponding to the N-terminal sequence GRDKYEPAAVSE (peptide 1-12) and against peptides corresponding to the C-terminal sequences IFVYDEVRKLIIRRR (peptide 991-1005) and RPGGWVEKETYY (peptide 1005-1016). These antibodies were purified by affinity chromatography on the respective peptide-Sepharose columns. Moreover, antibodies against the N-terminal dodecapeptide GRDKYEPAAVSE were obtained by affinity purification from heteroclonal antibodies against the alpha subunit of pork kidney Na+/K(+)-ATPase. These antibodies reacted with native as well as SDS-denaturated Na+/K(+)-ATPase. When the antibodies were used to probe the sidedness of the sequences in right-side-out vesicles of pig kidney microsomes, the N-terminal peptide 1-12 as well as the C-terminal peptides 991-1005 and 1005-1016 were found on the cytosolic side. Concanavalin A, however, which interacts with the beta subunit, a glycoprotein, reacted with the outside of right-side-out vesicles.

A monoclonal antibody against horse kidney (Na+ + K+)-ATPase inhibits sodium pump and E2K to E1 conversion of (Na+ + K+)-ATPase from outside of the cell membrane

Monoclonal antibodies against horse kidney outer medulla (Na+ + K+)-ATPase were prepared. One of these antibodies (M45-80), was identified as an IgM, recognized the alpha subunit of the enzyme. M45-80 had the following effects on horse kidney (Na+ + K+)-ATPase: (1) it inhibited the enzyme activity by 50% in 140 mM Na+ and by 80% in 8.3 mM Na+; (2) it increased the Na+ concentration necessary for half-maximal activation (K0.5 for Na+) from 12.0 to 57.6 mM, but did not affect K0.5 for K+; (3) it slightly increased the K+-dependent p-nitrophenylphosphatase (K-pNPPase) activity; (4) it inhibited phosphorylation of the enzyme with ATP by 30%, but did not affect the step of dephosphorylation; and (5) it enhanced the ouabain binding rate. These data are compatible with a stabilizing effect on the E2 form of (Na+ + K+)-ATPase. M45-80 was concluded to bind to the extracellular surface of the plasmamembrane, based on the following evidence: (1) M45-80 inhibited by 50% the ouabain-sensitive 86Rb+ uptake in human intact erythrocytes from outside of the cells; (2) the inhibition of (Na+ + K+)-ATPase activity in right-side-out vesicles of human erythrocytes was greater than that in inside-out vesicles; and (3) the fluorescence intensity due to FITC-labeled rabbit anti-mouse IgM that reacted with M45-80 bound to the right-side-out vesicles was much greater than that in the case of the inside-out vesicles.

Immunohistochemical localization of Na+, K+-ATPase in the choroid plexus

To determine if canine and rat choroid plexus Na+,K+-ATPase can be localized by immunoperoxidase staining after fixation and embedding, we prepared rabbit antiserum to purified canine kidney medulla Na+,K+-ATPase. When sodium dodecylsulfate polyacrylamide electrophoretic gels of purified canine kidney Na+,K+-ATPase and canine kidney microsomes were treated with antiserum followed by [125I]protein A and autoradiography, the canine microsomes and purified Na+,K+-ATPase showed a prominent radioactive band coincident with the alpha-, beta- and gamma-subunits of the purified canine kidney enzyme. When the rabbit immunoglobulin that was purified from the Na+,K+-ATPase antiserum through DEAE-cellulose ion exchange chromatography was used for immunoperoxidase staining of the choroid plexus fixed with Bouin's fixative, intense immunoreactive staining was present on the epithelial cells of both choroid plexuses but was not found in the tissue around the vessel. The staining was especially confined to apical surfaces of the epithelial cells. The same procedure was performed in the canine kidney, and immunostaining was obtained in the tubules where Baskin and Stahl described the enzyme localization. No staining was seen with pre-immune serum of the normal rabbit. We concluded that both the canine and rat choroid plexus are rich in Na+,K+-ATPase, which plays an important role in cerebrospinal fluid (CSF) secretion.

Immunochemical evidence for a transmembrane orientation of both the (Na+, K+)-ATPase subunits

Antibodies were raised against the large catalytic subunit (apparent Mr 96000) and the glycoprotein (apparent Mr 60000) of the sodium- and potassium-dependent adenosine triphosphatase [(Na+, K+)-ATPase] from Bufo marinus. The specificity of each antiserum was assessed by two-dimensional immunoelectrophoresis using toad kidney microsomes or the purified holoenzyme as a source of antigen and by indirect immunoprecipitation of detergent-solubilized (Na+, K+)-ATPase subunits from radioiodinated or biosynthetically labeled kidney holoenzyme, microsomes, or postnuclear supernatant. The anticatalytic subunit serum reacted exclusively with a 96000-dalton protein. The antiserum to the glycoprotein was rendered specific to this subunit by absorption with purified catalytic subunit. The two antisera were agglutinating and lytic in the presence of complement when toad erythrocytes were used as targets, indicating that antigenic determinants of both subunits were exposed on the cell surface. The specific reactivities with surface-exposed antigenic determinants of both subunits could be absorbed with toad red blood cells. Such absorbed antisera still reacted with detergent-treated or untreated kidney microsomes, revealing the presence of cytoplasmic and/or intramembranous antigenic sites. Our immunochemical data demonstrate that the glycoprotein subunit of (Na+, K+)-ATPase spans the lipid bilayer and confirm the transmembrane orientation of the catalytic subunit postulated from functional studies.

Differing sensitivities of Purkinje fibers and myocardium to inhibition of monovalent cation transport by digitalis

The extent of inhibition of monovalent cation active transport in Purkinje fibers and myocardium in response to toxic and inotropic doses of digitalis were studied in the dog to elucidate the factors underlying the different relative sensitivities of these tissues to the toxic arrhythmogenic effects of digitalis. Monovalent cation transport inhibition was assessed by measuring uptake of the K(+) analog Rb(+) in samples of myocardium and Purkinje fibers after in vitro ouabain exposure and after acute or chronic administration of digoxin in vivo. The active uptake of Rb+ was determined as the difference between total uptake and uptake in the presence of 1.0 mM ouabain. Mean active uptake of Rb(+) by Purkinje fibers from control hearts was 1.62+/-0.11 (SEM) nmol/mg wet wt per 15 min, significantly greater than the value of 0.49+/-0.05 for myocardium (P < 0.001). Concentration-effect curves for inhibition of monovalent cation active transport by in vitro exposure to graded concentrations of ouabain showed that the concentration for half-maximal inhibition of monovalent cation transport, IC(50), for Purkinje fiber transport was 0.4 muM, significantly less than the value of 1.4 muM for myocardial samples. Dogs were given toxic doses of digoxin (0.3 mg/kg i.v.). At onset of sustained ventricular tachycardia, they were killed and monovalent cation transport measured in myocardial and Purkinje fiber samples. Active Rb(+) uptake was inhibited by 44% in myocardial samples, whereas a significantly greater inhibition of 76% was noted in Purkinje fibers (P < 0.01). Similar data were obtained for both transmural myocardial biopsy samples and subendocardial trabecular samples obtained from regions adjacent to Purkinje fibers. Another group of dogs received a nontoxic dose of 0.02 mg/kg i.v. daily for 6 d. Myocardium showed a 17% reduction in Rb(+) active uptake compared to control animals receiving no drug, whereas Purkinje fiber transport was reduced in these dogs to a significantly greater extent averaging 44% (P < 0.001). Thus, both toxic and inotropic (non-toxic) doses of digoxin inhibited monovalent cation transport in Purkinje fibers to a greater extent than in myocardium. This difference in apparent sensitivity of monovalent cation transport to digoxin may contribute to the previously reported tendency of digitalis toxic arrhythmias to arise in Purkinje fibers.

Localization of the neurally mediated arrhythmogenic properties of digitalis

Available evidence suggests that the propensity of digitalis glycosides to produce cardiac arrhythmias is due in part to their neuroexictatory effects. We have performed experiments in cats which support the existence of a neurogenic component in the etiology of digitalis-induced ventricular arrhythmias. Our data further indicate that the locus of this neural effect lies within an area of the medulla 2 millimeters above to 2 millimeters below the obex. These findings, when considered with the effects of polar cardiac glycosides that do not cross the blood-brain barrier, suggest that the area postrema may be the site of neural activation.

Characteristics of antibody inhibition of rat kidney (Na+ -K+)-ATPase

Antibodies which were raised against highly purified membrane-bound (Na+ -K+)-ATPase from the outer medulla of rat kidneys inhibit the (Na+-K+)-ATPase activity up to 95%. The antibody inhibition is reversible. The time course of enzyme inhibition and reactivation is biphasic in semilogarithmic plots. In the purified membrane-bound (Na+-K+)-ATPase negative cooperativity was observed (a) for the ATP dependence of the (Na+ -K+)-ATPase activity (n = 0.86), (b) for the ATP binding to the enzyme (n = 0.58), and (c) for the ouabain inhibition of the (Na+ -K+)-ATPase activity (n = 0.77). By measuring the Na+ dependence of the (Na+ -K+)-ATPase reaction, a positive homotropic cooperativity (n = 1.67) was found. As reactivation of the antibody-inhibited enzyme proceeds very slowly (t0.5 = 5.2 hr), it was possible to measure characteristics of the antibody-(Na+ -K+)-ATPase complex: The antibodies exerted similar effects on the ATP dependence of the (Na+ -K+)-ATPase reaction and on the ATP binding of the enzyme. Vmax of the (Na+ -K+)-ATPase reaction and the number of ATP binding sites were reduced while K0.5 ATP for the (Na+ -K+)-ATPase activity and for the ATP binding were increased by the antibodies. The Hill coefficients for the ATP binding and for the ATP dependence of the enzyme activity were not significantly altered by the antibodies. The antibodies increased the K0.5 value for the Na+ stimulation of the (Na+ -K+)-ATPase activity, but they did not alter the homotropic interactions between the Na+-binding sites. The negative cooperativity which was observed for the ouabain inhibition of the (Na+ -K)-ATPase activity was abolished by the antibodies. The data are tentatively explained by the following model: The antibodies bind to the (Na+ -K+)-ATPase from the inner membrane side, reduce the ATP binding symmetrically at the ATP binding sites and reduce thereby also the (Na+ -K+)-ATPase activity of the enzyme. The antibodies may inhibit the ATP binding by a direct interaction or by means of a conformational change at the ATP binding sites. This may possibly also lead to the alteration of the Na+ dependence of the (Na+ -K+)-ATPase activity and to the observed alteration of the dose response to the ouabain inhibition.

Insulin effects on monovalent cation transport and Na-K-ATPase activity

The effects of insulin on monovalent cation transport and on Na-K-ATPase activity from intact cells, tissue homogenates, and purified enzyme of the avian salt gland were studied. Monovalent cation active transport, measured by ouabain-inhibitable 86Rb+ uptake, was significantly increased (21.9 +/- 7.3% SE) in tissue slices exposed to insulin (100 mU/ml) for 15 min. A small but significant (12.2 +/- 1.9%) increase in Na-K-ATPase activity was similarly observed after salt gland tissue slices were exposed to insulin. This increase in enzymatic activity did not occur when broken-cell homogenates were exposed to insulin. Purified preparations of Na-K-ATPase showed no insulin enhancement of activity either in the presence of optimal or less than fully activating Na+ and ATP concentrations. Na-K-ATPase activity was the same in detergent-activated homogenates of both control and insulin-treated slices, consistent with insulin activation of existing enzyme sites. These data support the hypothesis that at least part of the increase in monovalent cation active transport produced by insulin is related to enhanced Na-K-ATPase activity and indicate that the latter phenomenon is dependent on some components or properties of the intact cell.