Reaction of (Na,K)-ATPase with fluorescent maleimide derivatives. Probes for studying ATP site(s) function.

The fluorescent maleimide derivatives, 2-(4'-maleimidylanilino)naphthalene 6-sulfonic acid (Mal-ANS) and N-(1-pyrene)-maleimide (Mal-pyrene), both alkylate sulfhydryl groups on the alpha subunit of the (Na,K)-ATPase to inhibit (Na,K)-ATPase and p-nitrophenyl phosphatase activities and phosphoenzyme formation. Reaction of the enzyme with Mal-pyrene, but not with Mal-ANS, also inhibits MgPi- and Mg.ATP.Na-supported [3H]ouabain-binding to the enzyme. Mal-pyrene and Mal-ANS react, in part, with different sulfhydryl groups on the enzyme protein. On the average, the sulfhydryl groups which react with Mal-pyrene are located in a more shielded or hydrophobic environment than are those which react with Mal-ANS. It is the reaction of Mal-pyrene with sulfhydryl groups, which are not accessible to Mal-ANS, that results in the decreased [3H]ouabain-binding capacity of the (Na,K)-ATPase. The results indicate that phosphorylation of (Na,K)-ATPase is not required for Mg.ATP.Na-stimulated ouabain binding, and suggest that the ATP and sodium sites which modulate the interaction of ouabain with the (Na,K)-ATPase may be different from those which promote phosphorylation.

added to 0.1 mg of enzyme protein/&. Reactions of the (Na,K)-ATPase with Mal-ANS were carried out as described previously (12).
For flourescence polarization measurements, the polarizer accessory for the MPF series (Perkin-Elmer) was used. Uncorrected fluorescence polarization was measured using the equation p = (4, -IL)/ (41 + I d , where 41 and I A are the fluorescence intensities of parallel and perpendicular components of polarized light. Measurements of the collisional quenching of Mal-pyrene and Mal-ANS fluorescence were conducted by adding various concentrations of freshly prepared NaI to the sample cuvette and adding the same concentration of NaCl to the control cuvette. The ratio Fo/F was calculated from the initial fluorescence intensity (Fo) and the fluorescence intensity (F) measured at each concentration of NaI added to the probe-enzyme complex. The ratio was normalized by dividing by the Fo/F value obtained by adding the same concentration of NaCl to the probe-enzyme complex.
The amounts of Mal-pyrene and Mal-ANS bound to the (Na,K)-ATPase preparation were calculated assuming e3322 = 20,000 M" cm" (Mal-ANS) and = 40,000 M" cm" (Mal-pyrene). In order to quantitate the amount of probe bound to each of the protein subunits and the incorporation of probe into the lipids, 8 mg of Mal-ANS-or Mal-pyrene-labeled (Na,K)-ATPase were solubilized with 2% SDS in the presence of 25 m~ Tris-C1, pH 7.5, plus 10 mM 2-mercaptoethanol and applied to a Bio-Gel A-5m column (1.5 X 190 cm) equilibrated and eluted with 25 m~ Tris-C1, pH 7.4,0.1% SDS, 50 m~ NaCl, 1 m~ 2-mercapatoethanol, and 0.02% NaN3. The fractions containing the subunits were concentrated by lyophilization and then quantitated by measuring the absorbance at 343 or 322 nm, as described previously for Mal-ANS (12).
Other Methods-Lipid phosphorous was determined according to Bartlett (16). SDS-polyacrylamide gel electrophoresis was conducted as described by Laemmli (17) on 7.5% polyacrylamide gels. It has been reported (18,19) that the colorimetric protein assay described by Lowry et al. (20) overestimates the amount of (Na,K)-ATPase protein by approximately 40%. In addition, the molecular weights of the a and p subunits, as well as the number of each of these subunits comprising a functional unit of the (Na,K)-ATPase, are in question (21)(22)(23). For determining the relative levels of Mal-ANS and Malpyrene labeling of (Na,K)-ATPase in the present study, the enzyme and subunit protein concentrations have been estimated by the method of Lowry et al. (20), using bovine serum albumin (Sigma) as standard. Further, it has been assumed that the functional unit of (Na,K)-ATPase is a& (M, = 278,000), with M , ( a ) = 95,000 and M, (p) = 43,800.

RESULTS
Reaction of Probes with (Na,K)-ATPase-The addition of various concentrations of Mal-pyrene to 0.1 mg/ml of (Na,K)-ATPase resulted in a time-dependent increase in the fluorescence intensity at 375 nm, indicating incorporation of the probe into the enzyme preparation. With concentrations of Mal-pyrene lower than 5 p~, this time-dependent increase in the fluorescence intensity was proportional to the amount of Mal-pyrene incorporated into the enzyme preparation. The reaction of Mal-pyrene is similar to that described previously (12) for the reaction of Mal-ANS with this same enzyme preparation, except that the total incorporation of Mal-pyrene into the (Na,K)-ATPase Preparation was about 2.5-fold higher than that observed with Mal-ANS when the labeling reactions were carried out under identical conditions. T o determine whether the greater incorporation of Malpyrene was due to the specific labeling of the a subunit, as was observed previously with Mal-ANS (12), or was due in part to nonspecific incorporation of Mal-pyrene into either the enzyme proteins or the associated lipids, the following experiments, which are summarized in Table   I, were conducted: I) (Na,K)-ATPase (0.1 mg/ml) was incubated with 2mercaptoethanol prior to the addition of 10 p~ Malpyrene (line 5). The reaction was continued for 30 min and the enzyme was then washed twice by centrifugation and resuspension to remove unbound probe. Since the preincubation of the (Na,K)-ATPase with 2-mercaptoethanol prevents the alkylation of protein sulfhydryl groups by Mal-pyrene, the

TABLE I Quantitation and fluorescence polarization of Mal-pyrene incorporated into (Na,K)-ATPase and into extracted lipids under dicferent ligand conditions
(Na,K)-ATPase (0.1 mg/ml) or extracted lipids (0.1 pmol/ml) were incubated in 10 m~ Tris-C1, pH 7.4, and the additions shown, with varying concentrations of Mal-pyrene for 30 min at 22 "C. The enzyme in line 5 was preincubated with 0.15 m~ 2-mercaptoethanol prior to the addition of Mal-pyrene. The reactions were terminated by the addition of 0.15 m~ 2-mercaptoethanol, and unbound Mal-pyrene was removed by centrifugation and resuspension (lines [1][2][3][4][5] or by dialysis (line 6). In lines 7 and 8, 10 p~ Mal-pyrene plus 0.1 mM 2mercaptoethanol was added to 10 m~ Tris-C1 or to glycerol. Fluorescence polarization was measured at 24 "C as described under "Materials and Methods." results of this experiment indicated that 4.9 mol of Malpyrene/mol of (Na,K)-ATPase were nonspecifically incorporated into the enzyme preparation. The results of this experiment were identical with those obtained when Mal-pyrene was preincubated with a 5-fold excess of 2-mercaptoethanol prior to its addition to the enzyme suspension (data not shown). 11) Lipid vesicles (0.1 pmol of phospholipid/ml) which were prepared from lipids extracted from the (Na,K)-ATPase preparation by chloroform/methanol were incubated with 10 IJM Mal-pyrene for 30 min and then dialyzed overnight against 10 m~ Tris-C1, pH 7.4, a t 3 "C (line 6). Since this (Na,K)-ATPase preparation contains approximately 1 pmol of lipid phosphorus/mg of protein, the amount of Mal-pyrene associated with 1 pmol of lipid phosphorus represents the amount of Mal-pyrene associated with the lipid moiety of 1 mg of protein of the enzyme preparation. In both Experiments I and 11, 4 to 5 mol of Mal-pyrene were found to be associated nonspecifically with 1 mol of the (Na,K)-ATPase preparation, assuming that 1 mg of protein equals 3.6 nmol of (Na,K)- T o determine if a time-dependent hydrolysis of bound Malpyrene occurred, as reported previously by Lux and Gerard (24), the emission spectra of freshly prepared (Na,K)-ATPase-Mal-pyrene complex and an a-Mal-pyrene complex, which had been isolated from this labeled enzyme and stored for 2 months, were compared. The two emission spectra were virtually identical with respect to emission maxima and ratios of the peak heights (data not shown), indicating that hydrolysis of protein-bound Mal-pyrene had not taken place.
Location of Bound Probes-The limiting values of the fluorescence polarization of Mal-pyrene plus 2-mercaptoethanol, at room temperature, in buffer (n = 1 cp) and in glycerol (n = 1,000 cp) were 0.009 and 0.174, respectively (Table I).
The fluorescence polarization of the nonspecifically labeled by guest on March 24, 2020 http://www.jbc.org/ Downloaded from (Na,K)-ATPase or lipid vesicles, 0.049 and 0.091, respectively, is substantially lower than the fluorescence polarization (0.136) of the Mal-pyrene-enzyme complex which was formed in the absence of 2-mercaptoethanol. This indicates that the Mal-pyrene molecules which are nonspecifically associated with the enzyme preparation and/or intercalated into the lipid moiety have a higher degree of freedom of motion than do those molecules which are specifically bound to the enzyme proteins.
To determine the amounts of Mal-pyrene and Mal-ANS covalently bound to the (Na,K)-ATPase protein subunits, the labeled enzymes were solubilized with SDS and chromatographed on a Bio-Gel A-5m column to separate the a and / 3 subunits and the lipids (Fig. 1). Since both Mal-pyrene and Mal-ANS absorb in the 280 nm region, the peaks of relative absorbance at 280 nm shown in Fig fractions was estimated by the method of Lowry et al. (20), and the relative absorbance at 750 nm of the colorimetric assay for protein of these fractions is shown in Fig. 1B. The relative fluorescence of the fractions ( Fig. 1 0 reveals the presence of a large quantity of Mal-pyrene, but not Mal-ANS, in the lipid-containing fractions which elute behind the / 3 subunit. As shown in Table 11, approximately 50% of the Malpyrene which is associated with the (Na,K)-ATPase preparation is eluted with the lipid fraction. For both Mal-pyrene and Mal-ANS, most of the covalently bound probe is located on the a subunit of the enzyme. A small quantity of both probes (approximately 1 mol/mol) appears to be associated with the fractions containing the /3 subunit of the enzyme (Fig. 1). This observation appears to be inconsistent with our earlier conclusion (12) that Mal-ANS labeled only the a subunit of the (Na,K)-ATPase. The fluorescence intensity of the p subunit-containing fractions is so low with both probes that it can only be detected by monitoring the column fractions with a fluorescence spectrophotometer, as was done in Fig. 1 previously for the reaction of Mal-ANS with this enzyme, suggesting that the same M P site@) is regulating the interaction of the enzyme with both probes. The location of the additional six sulfhydryl groups on the (Na,K)-ATPase which are labeled by Mal-pyrene in the presence of 5 m~ MgC12 was investigated using steady state fluorescence polarization of the Mal-pyrene-enzyme complex. As shown in Table I alone. This suggests that the additional sulfhydryl groups, which are labeled in the presence of MgC12, may be located in a more fluid environment than those which are labeled in the absence of MgC12, thereby lowering the average polarization value.
It is also apparent from the data in Table  I, that the fluorescence polarization of the Mal-pyrene-(Na,K)-ATPase complex is highest when the probe concentration and, therefore, the number of sulfhydryl groups labeled is the lowest (line 4). With higher concentrations of Mal-pyrene (lines 2 and 3), the fluorescence polarization decreases indicating, on the average, a higher degree of freedom of rotation around the covalent bond. Since the relative quantities of covalently bound probe and nonspecifcally incorporated probe are equal for 2,10, and 30 PM Mal-pyrene (data not shown), the observed concentration dependence of the fluorescence polarization is probably not due to the amount of nonspecifcally incorporated Mal-pyrene. Rather, the high degree of fluorescence polarization at 2 PM Mal-pyrene suggests that those sulfhydryl groups which have the highest affinity for Mal-pyrene are located within either crevices or regions of the (Na,K)-ATPase protein where the bound probe molecules can be immobilized by hydrophobic interaction with protein.
The effects of other ligands on the rate of Mal-pyrene the Mal-pyrene molecules which are bound to the (Na,K)-ATPase are located in a more shielded or more hydrophobic environment than are the Mal-pyrene molecules dissolved in aqueous buffer or incorporated into lipids.
The collisional quenching of Mal-ANS fluorescence with NaI revealed a different pattern of quenching (Fig. 2B). The apparent collisional quenching constants for the Mal-ANSenzyme complex and for Mal-ANS dissolved in 10 m Tris-C1, pH 7.4, plus 1 rn2-mercaptoethanol are 0.28 M" and 0.40 M-', respectively. This indicates a similar degree of accessibility of iodide ions to Mal-ANS bound to the (Na,K)-ATPase proteins and to Mal-ANS dissolved in aqueous buffer. It is not possible to directly compare the collisional quenching constants of these two probes, due to the large differences in the fluorescence life times of Mal-pyrene (about 70 ns) and Mal-ANS (about 14 ns) (26). However, the data presented with respect to the quenching constants of each probe in buffer, relative to the constant for the same probe bound to the (Na,K)-ATPase preparation, suggest that the Mal-pyrene molecules which are bound to the enzyme are in a more hydrophobic environment than are the Mal-ANS molecules.
Effect of Ligands-The rate of reaction of Mal-pyrene with (Na,K)-ATPase, as measured by an increase in the fluorescence intensity at 375 nm, is influenced by the ligands present in the reaction medium in essentially the same manner as that reported previously for Mal-ANS (12). The apparent binding Labeling of (Na,K)-ATPase Sulfiydryl Groups reaction with the (Na,K)-ATPase are illustrated in Fig. 3 and are essentially the same as those observed previously with Mal-ANS (12). As with Mal-ANS, comparison of the rates of fluorescence enhancement with 5 m~ MgC12, and 3 mM ATP (Curve A), 2 m~ MgC12 (Curve D ) , and 5 m~ MgC12, 3 m~ ATP, and 100 m~ NaCl (Curve C ) indicates that some of the sulfhydryl groups of the E-ATP form of the enzyme, but not of the Ez.P form, are protected from reaction with Malpyrene.
Effects ofMal-pyrene and Mal-ANS on EnzymicActivity-The effects of Mal-pyrene alkylation of the (Na,K)-ATPase are, in many respects, very similar to the effects observed previously with Mal-ANS (12). The reaction of both probes resulted in a time-and concentration-dependent loss of (Na,K)-ATPase activity and a parallel reduction in the steady state levels of phosphoenzyme formation and (Na,K)-ATPase activity (12) (Fig. 4). In addition, the loss of (Na,K)-ATPase activity in both cases was multiphasic, indicating that the loss of activity is due to the alkylation of multiple classes of The effects of Mal-pyrene and Mal-ANS reaction on the potassium-dependent p-nitrophenyl phosphatase activity are not the same. With Mal-pyrene, the loss of phosphatase activity paralleled the loss of (Na,K)-ATPase activity and the steady state level of phosphoenzyme formation (Fig. 4). As reported previously (E), the loss of (Na,K)-ATPase and phosphatase activities with Mal-ANS was parallel only to a level of approximately 50% inhibition, with the remaining phosphatase activity being much more resistant to inhibition than was the (Na,K)-ATPase activity.
The effects of Mal-pyrene and Mal-ANS on the interaction of ouabain with the (Na,K)-ATPase are markedly different. Alkylation of the (Na,K)-ATPase with Mal-pyrene resulted in a significant decrease in the [3H]ouabain-binding capacity of the enzyme. When the binding assay was carried out in the presence of 5 m~ MgC12,5 mM ATP, and 100 m~ NaCl (Mg. ATP-Na), the loss of [3H]ouabain-binding sites with increasing labeling by Mal-pyrene was almost parallel to the loss of (Na,K)-ATPase activity (Fig. 5A). Binding in the presence of sulfhydryl groups.  5 m~ MgCg and 5 m~ inorganic phosphate (Mg -Pi) was also reduced, but to a lesser extent than was the Mg.ATP.Nasupported binding. Nonspecific labeling of the enzyme preparation with Mal-pyrene as described above (Experiment I and line 5 of Table I) had no effect on either (Na,K)-.4TPase activity or the [3H]ouabain-binding capacity of the enzyme.
In contrast, the effects of MalANS alkylation on ouabain binding are much less apparent (Fig. 5B). We reported previously (12) that the ouabain-binding capacity of the (Na,K)-ATPase, measured in the presence of MgClz and inorganic phosphate, was only slightly reduced by Mal-ANS. We have now determined that Mal-ANS alkylation also causes only a slight reduction (approximately 14%, Fig. 5B) in the level of Mg -ATP -Na-supported ouabain binding. This is particularly interesting because the (Na,K)-ATPase activity of this Mal-ANS-labeled enzyme was 81% inhibited. Since the inhibition of phosphoenzyme formation from ATP parallels the inhibition of (Na,K)-ATPase activity (Fig. 8 of Ref. 12 and line 2 of Table IV), it appears that a large proportion of the Mal-ANSenzyme complex which cannot be phosphorylated by ATP in the presence of MgClZ and NaCl is still capable of binding ouabain in the presence of MgC12, NaC1, and ATP. To verify that the above measured equilibrium levels of C3H]ouabain binding to the Mal-ANS-labeled (Na,K)-ATPase represent inorganic phosphate-and ATP-stimulated binding, the rates of binding with various ligands were determined. As shown in  Table 111, although the rates of binding to the Mal-ANSenzyme complex are consistently two times slower than those to the control enzyme, the degrees of stimulation of the rates of binding by inorganic phosphate, ATP, and ATP plus sodium were the same for both control and Mal-ANS-labeled (Na,K)-ATPase. It might be argued that the binding measured in the presence of added magnesium and ATP was due to the presence or the generation of inorganic phosphate. It should be noted, however, that NaCl decreases the rate of ouabain binding in the presence of magnesium and inorganic phosphate. For example, in the presence of 5 m~ MgC12 and 5 mM inorganic phosphate the addition of 100 m~ NaCl increases the tlI2 for binding to control enzyme from 1.6 to 38 min. If the concentration of inorganic phosphate is lowered to 5 p~ (2.5 times the concentration of inorganic phosphate contamination present in the 80 PM ATP used here), the addition of 100 mM NaCl essentially prevents the binding of ouabain (data not shown). Since the rate of binding to the Mal-ANSlabeled (Na,K)-ATPase is six times faster in the presence of Mg. ATP. Na than it is with Mg. ATP (Table 111), it is apparent that the interaction of ouabain with the Mal-ANS-enzyme complex is regulated by ATP and by ATP + sodium, even though this modified enzyme cannot be phosphorylated by ATP.
As described above (Fig. 2), the collisional quenching of the fluorescence of the Mal-ANS-and Mal-pyrene-labeled (Na,K)-ATPase fractions indicated that at least some of the sulfhydryl groups which are labeled by Mal-pyrene are different from those which are labeled by Mal-ANS. To determine if it is the labeling of different sulfhydryl groups by Malpyrene that is responsible for the greater inhibition of ouabain binding, (Na,K)-ATPase was labeled sequentially with Mal-ANS and Mal-pyrene. The concentrations of the probes and the incubation times were selected so that the degree of inhibition of (Na,K)-ATPase activity was equivalent with both Mal-ANS and Mal-pyrene. As shown in Table IV (line   2) the Mal-ANS-treated enzyme fraction appears to consist of approximately 10% native enzyme, 17% totally inactive enzyme (with respect to ATP hydrolysis and ouabain binding), and 83% that has been modified such that ATP and inorganic phosphate are capable of supporting ouabain binding even though the enzyme cannot be phosphorylated by ATP. (Na,K)-ATPase that has been labeled only with Mal-pyrene (line 3) appears to contain 10% native enzyme, 62% totally inactive enzyme, and approximately 27% (38-11%) modified enzyme which can bind ouabain in the presence of magnesium    Table IV, which were carried out with varying concentrations of Mal-ANS and Mal-pyrene and slightly different incubation times revealed that Mal-ANS labels approximately 4 Sulfhydryl groups/a subunit, and that an additional 3-5 sulfhydryl groups/a subunit are labeled when the Mal-ANSenzyme complex is exposed to Mal-pyrene. These results indicate that it is the alkylation of additional sulfhydryl groups on the (Y subunit, which are not accessible to Mal-ANS, that is responsible for the significant inhibition of [3HJouabain binding by Mal-pyrene. Since preincubation of the (Na,K)-ATPase with ouabain altered neither the rates nor the apparent magnitude of Malpyrene reaction with the enzyme (Fig. 3), these additional Mal-pyrene-labeled sulfhydryl groups would not appear to be located within the ouabain binding site. Furthermore, in separate experiments (Na,K)-ATPase was preincubated with 5 p~ ouabagenin in the presence of Mg.Pi and Mg ATP -Na binding conditions, to protect the ouabain binding site from modification, and then exposed to Mal-pyrene. The labeled enzyme was washed by repeated centrifugation and resuspension to remove unbound Mal-pyrene and to dissociate ouabagenine from the enzyme. The amount of Mal-pyrene which bound to the (Na,K)-ATPase, the level of [3H]ouabain binding and the (Na,K)-ATPase activity were the same, whether or not the enzyme had been incubated with ouabagenin before and during the exposure to Mal-pyrene (data not shown).

DISCUSSION
The primary observations of this study are: 1) that the pyrene and ANS derivatives of maleimide react, in part, with different sulfhydryl groups on the (Y subunit, and have markedly different effects on the interaction of ouabain with the (Na,K)-ATPase; and 2) that phosphorylation of the (Na,K)-ATPase is not a prerequisite for ATP plus sodium-stimulated [3H]ouabain binding to the enzyme.
Both Mal-ANS and Mal-pyrene alkylate sulfhydryls on the (Y subunit which are essential for (Na,K)-ATPase activity, potassium-dependent phosphatase activity and phosphoenzyme formation. However, under the conditions used in the present study, only Mal-pyrene alkylates sulfhydryl groups which are essential for the binding of ouabain to its receptor site on the enzyme. From these experiments, it is not possible to determine either the location or the number of these essential sulfhydryl groups, but the results of the doublelabeling experiments, in which (Na,K)-ATPase was reacted first with Mal-ANS and then with Mal-pyrene, suggest that 5 or less sulfhydryls/a subunit are involved. Since pretreatment of the enzyme with ouabain or ouabagenin does not decrease the rate or the magnitude of Mal-pyrene reaction with (Na,K)-ATPase, nor does that pretreatment protect the enzyme against inactivation by Mal-pyrene, it seems reasonable to assume that the sulfXydry1 groups which are essential for ouabain binding are not located within the ouabain binding site of the (Na,K)-ATPase. The results do suggest that the interaction of ouabain with the (Na,K)-ATPase is regulated by a distinct region($ of the enzyme protein containing sdfhy-dry1 groups which are accessible to Mal-pyrene, but not to Mal-ANS.
On the average, the sulfhydryl groups which are alkylated by Mal-pyrene appear to be located in more shielded or more