photochemical labeling and inhibition of na,k-atpase by 2-azido

0 1994 by The American Society for Biochemistry and Molecular Biology, Inc. THE JOURNAL OF B I O ~ I C A L CHEMISTRY Vol. 269, No. 9, Issue of March 4, pp. 6558-6565, 1994

Prlnted in U.S.A.

Photochemical Labeling and Inhibition of Na,K-ATPase by 2-Azido-ATP IDENTIFICATION OF AN AMINO ACID LOCATED WITHIN THE ATP BINDING SITE*

(Received for publication, October 8, 1993, and in revised form, November 22, 1993)

Chinh M. ”ran$, Edward E. HustonSO, and Robert A. FarleyH II From the Wepartment of Physiology and Biophysics and the llDepartment of Biochemistry, University of Southern California School of Medicine, Los Angeles, California 90033

2-Azido-ATP (2-Ns-ATP) was investigated as a reagent for the identification of amino acids located within the catalytic ATP binding site of Na,K-ATPase. The enzymatic activity of Na,K-ATPase was inhibited up to 50% by 2-NS-ATP = 5-10 p ~ ) after irradiation with ultraviolet light, and inhibition was prevented by 0.2 m~ ATP. The binding of ATP to Na,K-ATPase (KO = 0.1 p ~ ) was inhibited competitively by 2-N3-ATP. [a-32P]2-N3-ATP labels the a subunit of Na,K-ATPase, and the stoichiometry of covalent ATP-protectable incorporation of the probe into the protein is approximately equal to the stoichiometry of high-affinity binding of ATP to the Na,K- ATPase. 2-N3-ATP is also hydrolyzed by Na,K-ATPase as a substrate. From these data, it is concluded that 2-N3- ATP photochemically labels the Na,K-ATPase from within the catalytic ATP site on the protein. Trypsin digestion of Na,K-ATPase after photochemical labeling with [a-32P]2-N3-ATP generated a large 30-kDa fragment containing the radiolabeled nucleotide. This fragment was resistant to further cleavage by trypsin, but it could be digested further after denaturation in urea. High pressure liquid chromatography separation of tryptic peptides from the 30-kDa fragment and subsequent amino acid sequence analysis of the radiolabeled pep tides identified the region between His496 and A r g B ’ O of the Na,K-ATPase a subunit as the region labeled by [a-32P]2-N3-ATP. Glyso2 was absent from all sequences of the radiolabeled peptides from this region, consistent with the derivatization of this amino acid by 2-N3-ATP and localization of Gly602 within the ATP binding site.

P-type ion pumps comprise a family of structurally related proteins that directly couple the energy released from ATP hydrolysis to the translocation of ions across cell membranes. Although the mechanism of ATP-coupled ion transport is not completely understood, it is known that the proteins bind ATP and that the y-phosphate of ATP is then transferred to an aspartate residue on the protein to form a phosphoenzyme intermediate. The phosphoenzyme intermediate is hydrolyzed, and phosphate is released from the protein as the reaction cycle is completed. Both the formation of the phosphoenzyme and its

* This work was supported by Public Health Service Grant GM28673, by the American Heart Association, Greater Los Angeles Affiliate Fel- lowship 885, and by the American Heart Association, Greater Los An- geles Affiliate Investigative Group Award IG 809. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “aduertisernent” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

5 Present address: Sterling Winthrop, Inc., Collegeville, PA 19426- 0900.

I( To whom correspondence should be addressed: Dept. of Physiology and Biophysics, USC School of Medicine, 2025 Zonal Ave., Los Angeles, CA 90033.

subsequent hydrolysis are associated with conformational changes in the protein that are thought to be linked to the movement of the ions across the cell membranes (reviewed in Ref. 1).

Identification of amino acids that are located within the ATP binding site of P-type ion pumps is essential to understand the structural basis for ATP-coupled ion transport. Several chemical reagents with different structural characteristics have been used in attempts to identify amino acids within the nucleotide binding sites of these proteins, and, in general, two regions of primary structure that are separated by about 200 amino acids have been labeled by these reagents. In the Na,K-ATPase,’ these regions lie between amino acids 470 and 501 and between amino acids 710 and 720. In the first group of amino acids, it has been observed that LysBol is labeled by FITC (2, 3) and SITS (4), Lys480 is labeled by adenosine diphosphopyridoxal(5), pyridoxal phosphate (51, and FITC (61, and amino acids between 470 and 495 are labeled by 8-N3-ATP (7). In the second group, p-(fluorosulfony1)benzoyladenosine has been reported to label Na,K-ATPase at Lys719 (8), and ClR-ATP labels Asp710 and possibly Asp714 (9). Although inhibition of Na,K-ATPase enzymatic activity by each of the reagents is prevented by ATP, several of the inhibitors bear little structural resemblance to ATP, and inhibition may result from interactions outside of the nucleotide binding site. Among the reagents listed above, only 8-N3-ATP is a substrate for Na,K-ATPase.

In the experiments reported here, an arylazide derivative of ATP, 2-N3-ATP, has been used to photochemically label the nucleotide binding site of Na,K-ATPase purified from canine renal medulla. As was observed for 8-N3-ATP (101, 2-NS-ATP is used by the enzyme as a substrate. Thus, it is likely that the probe labels the enzyme from within the active site. The identification of Gly502 as the amino acid that is labeled by 2-N3- ATP indicates that amino acids in this region of the protein are likely to interact with the purine ring of the nucleotide.

EXPERIMENTAL PROCEDURES MateriaZs-2-Chloroadenosine, phosphorous oxychloride, and

DEAE-Sephadex were purchased from Sigma. Anhydrous hydrazine, p-toluenesulfonic acid, ethyl orthoformate, and trichloroacetonitrile were from Aldrich, and cellulose GF-TLC plates and triethylamine were from J. T. Baker. The triethylamine was redistilled under reduced pressure before use. H332P04 and [2,8-3HlATP were from Amersham Corp.

from Worthington and was purified by HPLC before use. Na,K-ATPase L-(’Ibsylamido 2-pheny1)ethyl chloromethyl ketone-treated trypsin was

The abbreviations used are: Na,K-ATPase, sodium- and potassium- dependent adenosine triphosphatase (E.C. 3.6.1.37); FITC, fluorescein- 5’-isothiocyanate; SITS, 4-acetoamido-4‘-isothiocyanostilbene-2.2’-di-

8-azido-adenosine triphosphate; ClR-ATP, y-(4-N-2-chloroethyl-N- sulfonic acid; 2-N3-ATP, 2-azido-adenosine triphosphate; 8-N3-ATP,

methy1amino)benzylamide-adenosine triphosphate; HPLC, high pressure liquid chromatography; PAGE, polyacrylamide gel electrophoresis.

6558

2-Azido-ATP Labeling of Na,K-ATPase 6559

was purified from the outer medulla of dog kidneys by the method of Jorgensen (11).

Synthesis of 2-N3-ATP-The synthesis of 2-N3-ATP was done by com- bining selected steps from the methods of Schaeffer and Thomas (12), Michelson (13), Yoshikawa et al. (14), Tomasz (151, Czarnecki (16), and Boulay et al. (17) and consists of three steps involving conversion of 2-chloroadenosine to 2-hydrazinoadenosine, converting 2-hydrazinoadenosine to 2-azidoadenosine and phosphorylation of the nucleoside first to azido-AMP and then from azido-AMP to azido-ATP. A typical yield of 2-N,-ATP was 15-20% compared with 2-chloroadenosine. The progress of the reactions were monitored by thin-layer chromatography.

Synthesis of [n-3'P]2-N,,-ATP-The synthesis of [ c x - ~ ~ P J ~ - N ~ - A M P was performed according to Symons (18). The conversion of the mono- phosphate to the triphosphate species was performed using adenylate kinase (myokinase) with glyceraldehyde 3-phosphate as substrate, as a modification of a method described by Melese and Boyer (19). Potassium phosphate was used to phosphorylate the [a-3zP12-N,-AMP to [~ r -~~P12- N,-ATP. The identity of the product was verified by comparing its RF with nonradioactive 2-N3-ATP after TLC and autoradiography.

Photolysis of Na,K-ATPase with 2-N:,-ATP-The final conditions for photolysis were as follows. Protein concentration was 0.2 mg/ml in a

buffer was 10 m~ Tris-HCI, pH 7.0, 10 mM NaCI, 2.5 m~ MgCl,, 0.2 m~ final volume of 5 ml in a 6-well Costar cell culture plate. The photolysis

Na2EDTA, 0.1 m~ glutathione. The photon source was two $-watt UV lamps (W Products, San Gabriel, CA) with peak emission at 302 nm, set up about 30 cm from the samples. Pyrex culture plates were used to filter wavelengths shorter than 290 nm. The incident energy was about 400 pW/cm2 on the samples. The sample plates were set on an orbital shaker and were shaken gently during photolysis. The photolysis time was 30 min because longer times did not increase the ATP protectable inhibition and did increase the nonspecific inhibition. ATP-protectable inhibition is defined as the difference in inhibition between identical samples that contained the same amount of 2-N,-ATP, either with or without 0.2 mM ATP. 2-N3-ATP concentrations were between 1 and 50 PM.

Na,K-ATPase Activity Assays-After photolysis, 1 0 4 aliquots from each sample were assayed for 15 min a t 37 "C for remaining activity. The assay buffer contained 100 mM Tris-HCI, pH 7.4, 100 mM NaCl, 5 mM KC1,2.5 mM MgCl,, 1 mM Na,EDTA, 3 mM ATP, with and without 1 mM ouabain. The reactions were initiated by addition of 10 pl of enzyme to 1 ml of the incubation buffer and were stopped by addition of 0.2 ml of 30% trichloroacetic acid. The samples were incubated on an ice slurry for 30 min, and then 0.5 ml of the reaction buffer from each sample were used to assay for the inorganic phosphate (20). For p-nitrophenylphosphate hydrolysis, 20 pl from each photolyzed sample were used to assay

buffer contained 50 mM Tris. HC1, pH 7.4, 10 mM MgCI,, 100 mM KCI, 5 for remaining ouabain-sensitive p-NPPase activity. The incubation

mM EDTA (free acid), and 5 mM p-nitrophenylphosphate, with and without 1 mM ouabain. The reaction was initiated by addition of 20 pl of enzyme solution to 2 ml of incubation solution, and after 30 min at room temperature, 4 ml of 1 N NaOH, 0.1% Triton X-100 was added to each tube to quench the reaction. Absorbance a t 410 nm was measured, and the difference between samples with and without 1 mM ouabain was used to calculate the remaining activity. For 2-N3-ATP hydrolysis, purified Na,K-ATPase was suspended in buffer without nucleotides. The reaction was started by addition of different concentrations of ATP or 2-N3-ATP. The final composition of the assay buffer was 20 mM HEPES- triethylamine, pH 7.4, 1.5 mM phosphoenol pyruvate, 125 mM NaCI, 25 mM KC1, 2 mM MgCl,, 0.125 mM Na,EDTA, 1.25 mM dithiothreitol, 0.315 mg/ml P-NADH (reduced form), 25 pg/ml pyruvate kinase, 25 pg/ml lactate dehydrogenase. Activity was measured as the change in absorbance a t 340 nm.

Equilibrium Binding of ATP-Purified Na,K-ATPase samples of 0.1 mg each were suspended in 1 ml of 20 mM Tris, HCI, pH 7.4,3 m~ EDTA (free acid), which also contained 50 pCi of [2,8-,H]ATP and Tris-ATP of

Tris-ATP to obtain the nonspecific binding of [2,8-"H]ATP. Identical sets various concentrations from 0.1 to 5 PM. One sample contained 5 mM

of samples were set up, and 2-N3-ATP was added to final concentrations of either 10 or 20 p ~ . Each measurement was made in duplicate. Samples were kept on ice for 5 min and were centrifuged in a Ty 65 rotor a t 50,000 rpm for 60 min to recover the enzyme pellets for assaying. The

dry with cotton tip applicators. The pellets were then dissolved in 200 supernatants were removed by aspiration, and the pellets were blotted

pl of 1% SDS, transferred to scintillation vials, and counted with 3 ml of scintillant 3a70b. Counts as disintegrations/min from the 5 mM ATP samples were considered as nonspecific binding and were substracted from all other samples. The remaining counts were used to calculate the amount of ATP bound to the enzyme. The results were calculated as

nanomoles of ATP bound per milligram of enzyme. Stoichiometry of Labeling-Purified Na,K-ATPase was photolyzed in

the following conditions. Final volume of each sample was 5 ml with a protein concentration of 0.2 mg/ml, with different concentrations of [(Y-"~P]~-N,-ATP (specific activity, 100 pCi/pmol) in the presence and absence of 0.2 mM ATP. The photolysis buffer was the same as described earlier. Samples were photolyzed for 30 min at 300 pW/cm2, and 10 pl from each sample was used to measure remaining Na,K-ATPase activity as described above. The remaining activities of samples containing 0.2 mM ATP were subtracted from corresponding samples that had 2-N3-ATP only to give the amount of ATP-protectable inhibition. After removing the aliquot for ATPase measurement, the remainder of each sample was centrifuged in a Ty 65 rotor at 45,000 rpm for 90 min. The supernatants were discarded, and the pellets were resuspended in 25 m~ imidazole-HC1, pH 7.4, 1 m~ Na2EDTA to a final concentration of 2 mg/ml. The resuspended pellets then were counted for radioactivity as Cerenkov counting. A known volume of the [a-32PJ2-N,-ATP used in the experiment was counted along with the samples to calculate the exact specific activity. The counts as countdmin from samples with 0.2 m~ ATP were subtracted from the corresponding samples of the same [a-"P]2-N3-ATP concentrations, and remaining countdmin were used to calculate the ATP-protectable labeling as nmdmg. About 100 pg from each sample were also analyzed by SDS-polyacrylamide gel electrophoresis (21). After the gel was stained and destained, the a-subunit bands were excised from the gel and counted for radioactivity. The countsimin collected from the gel slices were used to calculate the amount of [a-32P12-N,-ATP, as described for the Na,K-ATPase pellets.

Photolabeling of the Enzyme-10-15 mg of Na,K-ATPase was labeled in 6-well Costar culture plates as described above with 50 p of [a-3zP12- N,-ATP. The specific activity of the azido nucleotide was between 100 and 300 pCUpmo1. 5 mg of the same enzyme preparation was labeled with 0.2 mM ATP to estimate the nonspecific inhibition by 2-N3-ATP. After photolysis, 10 pl from each sample were assayed for remaining activity using a coupled spectrophotometric assay. Samples were centrifuged in a Ti 70 rotor a t 48,000 rpm for 90 min. The supernatants were discarded and the pellets were resuspended in 25 mM imidazole- HCI, pH 7 .4 , l m~ Na,EDTA to about 2 mg/ml and were centrifuged one more time. The pellets were collected, resuspended in the same buffer, and stored in -85 "C until needed.

Digestion of Labeled Enzyme for HPLC Analysis-About 10-15 mg of labeled enzyme was suspended in 25 m~ imidazole-HC1, pH 7.4, 1 mM Na,EDTA to a concentration of 2 mg/ml and was digested with trypsin at a ratio of 1 5 0 for 3 h at 37 "C. The membranes were separated from the soluble peptides by centrifugation in a Ti 70 rotor for 90 min at 48,000 rpm. The supernatant was collected, and the pellet was discarded. Solid urea was added to the supernatant to a final concentration of 5 M. The supernatant was incubated with urea for 3 h a t 37 "C and then was diluted with 25 mM imidazole-HC1, pH 7.4, 1 m~ Na,EDTA to a final urea concentration of 2.5 M. Trypsin was added at a ratio of 150 (w:w) with respect to the original amount of Na,K-ATPase. The sample was incubated a t 37 "C for 3 h, and the reaction was stopped by immer- sion in a boiling water bath for 5 min. The sample was diluted 10-fold with 20 mM phosphoric acid-NaOH, pH 7.4, and was loaded on a 30-g Sep-Vac Pak C18 cartridge (Millipore Corp.). The cartridge was washed with 100 ml of 20 mM phosphoric acid-NaOH, pH 7.4, and was extracted with 60 ml of 10 mM phosphoric acid-NaOH, pH 7.4 in 80% acetonitrile. The organic extract was collected, and the acetonitrile was removed by

material was stored at -85 "C until further analysis by HPLC. Typi- rotary evaporation under reduced pressure. The remaining aqueous

cally, about 50% of the applied radioactivity did not bind to the cartridge, and about 60% of the bound radioactivity was extracted with the phosphoric acid-acetonitrile solvent. About 20% of the applied radioactivity was irreversibly bound to the cartridge.

HPLC Analysis-The aqueous extract from the Sep-Vac Pak cartridge was injected to a Cla column for analysis. The solvents and gradient used were as follows. Solvent A was 20 mM H3PO4-NaOH, pH 6.5, solvent B was 70% acetonitrile in water; the gradient was from 0%-100% solvent B in 90 min at a flow rate of 1.5 mVmin (22). Fractions of 2 min each (3 ml) were collected using a Gilson fraction collector. The absorbance was monitored simultaneously with a Hitachi 100-40 single wavelength detector a t 280 nm and an LKB 2140 photodiode array detector. Fractions were later counted for radioactivity (Cerenkov counting). Pools of radioactive fractions were collected and later separated on a C, reverse phase column (25 cm x 4.5 mm, Vydac) with a solvent system consisting of solvent A (10 m~ H,PO,-KOH, pH 2.55) and solvent B (70% acetonitrile in water) (22). A 90-min linear gradient of 0 4 0 % of solvent B was used with a flow rate of 1.5 mumin. 3-ml fractions were collected, and radioactive peaks were pooled. Final

6560 2-Azido-ATP Labeling of Na,K-ATPase

100 I I I 1

0 5 10 15 20 25 0.00 0.60 1.20 1 .so 2.40 3.00

2N,ATP (pM)

FIG. 1. Photoinhibition of dog kidney N-K-ATPase by 2-N,- ATP. Purified Na,K-ATPase was suspended in 10 m~ Tris-HC1, pH 7.0, 10 nm NaC1, 2.5 m~ MgC12, 0.2 m~ Na2EDTA, 0.1 m~ glutathione in reduced form. 2-N3-ATP alone, or both 2-N3-ATP and 0.2 m~ ATP, and buffer were added to each well to final concentrations of 1,2,3,5,8, 15, and 20 p~ 2-N3-ATP in a final volume of 5 ml for each sample. The protein concentration of the samples was 0.2 mg/ml. Samples were shielded with Pyrex culture dishes and were photolyzed at 400 pW/cm2 for 30 min at room temperature with two 8-W W lamps. After photolysis, the remaining Na,K-ATPase activities were assayed as described in the text. The results from 6 separate experiments are shown.

HPLC separation was done on a CIB column with 0.05% trifluoroacetic acid as solvent A and 0.05% trifluoroacetic acid in 70% acetonitrile in water as solvent B. The gradient was the same as one for the C, separation. Fractions of 0.5 min were collected to increase the resolution of collected radioactive peaks. After being counted for radioactivity, peaks of interest were stored in -85 “C until needed for sequencing.

RESULTS Photoinhibition of Na,K-ATPase Actiuity-Reaction condi-

tions for the photochemical labeling of Na,K-ATPase by 2-N3- ATP were determined by systematically varying pH, photolysis time, incident ultraviolet energy, protein concentration, and temperature. To minimize the damage to the protein by W light and to avoid inner filter effects due to the absorption of incident energy by the purine of either 2-N3-ATP or ATP, the photolysis experiments were done using incident wavelengths greater than 300 nm. The criteria used to evaluate the effec- tiveness of the photolysis conditions were the extent of protection from inactivation in the presence of ATP and the extent of nonspecific inactivation of activity by UV light in the absence of 2-N3-ATP. Conditions under which the protection by ATP was the greatest and the nonspecific inactivation was the least are described under “Experimental Procedures” and were used in all studies unless otherwise noted. The irreversible inhibition of ATP hydrolysis by different concentrations of 2-N3-ATP and the protection by ATP from azido nucleotide-dependent inactivation are shown in Fig. 1. The activity of the enzyme is inhibited to 50% of the initial activity by about 10 p~ 2-N3-ATP under these conditions. Inhibition of the hydrolysis of p-nitrophenylphosphate with a of 5-10 p~ was also observed (data not shown). Longer irradiation times resulted in greater inhibition of activity; however, the nonspecific azido nucleotide- independent inactivation of the enzyme also increased.

Znhibition of ATP Binding by 2-N3-ATP-Protection by ATP against enzyme inhibition due to photoactivation of 2-NB-ATP is due to competition for binding sites on the Na,K-ATPase between the azido nucleotide and ATP. This was apparent in assays of equilibrium binding of I3H1ATP to Na,K-ATPase in the presence and absence of Z-N3-ATP (Fig. 2). ATP binds to a high-affinity site in this preparation of Na,K-ATPase with a KD of about 0.4 p~ and a B,, of 2.4 nmol/mg. In the presence of 10

Bound (nmol/mg)

FIG. 2. Equilibrium binding of ATP in the presence and absence of 2-Ns-ATP. Samples of purified Na,K-ATPase, 1 ml each of 0.1 mg/ml in 20 m~ Tris-HC1, pH 7.4,3 m~ EDTA, 50 pCi of [2,8- 3HIATP, and various concentrations of Tris-ATP from 0.1 to 5 p~ in the presence (open triangles) and absence (filled triangles) of 20 p~ 2-N3-ATP. Samples were incubated on ice for 5 min, and the membranes were separated from unbound ligands by centrifugation in a T y 65 rotor at 50,000 rpm for 1 h. The supernatants were discarded, and the pellets were blotted dry with cotton tip applicators and dissolved in 0.2 ml of

5 m~ Tris-ATP were considered background counts and subtracted from 1% SDS for radioactivity counting. The counts from samples containing

all samples. The remaining counts were used to calculate the amount of nucleotide bound as nmoVmg and were plotted as a Scatchard plot. The apparent KD values in the presence (0.94 p ~ ) and absence (0.47 p ~ ) of 2-N3-ATP were extracted from the plot and used to calculate the K, for 2-N3-ATP according to the equation

(Eq. 1)

The E,, of ATP binding to the enzyme in the experiments was 2.4 nmoVmg and the K, for 2-N3-ATP inhibition is about 20 p ~ . Another experiment was done in the presence of 10 p~ 2-N3-ATP, and the calculated KO and K, values were the same as those reported above.

p~ (data not shown) or 20 p~ 2-N3-ATP, the apparent KO for ATP binding is increased but the B,, remains unchanged. The Ki for 2-N3-ATP inhibition of ATP binding was calculated from the equation KD(app) = ((KIJK).[Il) + KD, where KD and K D ( ~ ~ ~ ) are obtained from Scatchard plots in the absence or presence of 2-N3-ATP (I), respectively.

Hydrolysis of 2-N3-ATP by Na,K-ATPase-The competitive interaction between ATP and 2-N3-ATP that was observed in the equilibrium binding assay suggests that the two nucleotides bind to the protein at the same site. Mutually exclusive binding at different sites, however, cannot be excluded by this assay, and to determine whether the 2-N3-ATP is binding at the high-affinity ATP site, the ability of Na,K-ATPase to use 2-N3- ATP as a substrate was investigated. The hydrolysis of different concentrations of Z-N3-ATP by Na,K-ATPase is shown in Fig. 3. The Michaelis-Menten equation was fit to the data, and a V,, of 0.88 pmols ATP/midmg protein and a K, of 50 p~ were obtained from the fit. The V,, for hydrolysis of ATP by this sample was 25 pmol ATP/min/mg protein. The hydrolysis of 2-N3-ATP was inhibited 90-95% by 0.2 m~ ouabain, and the rate of hydrolysis of Z-N3-ATP was not affected by increased concentrations of lactate dehydrogenase and pyruvate kinase used in the assay. In a separate assay (Fig. 3, inset ), 2-&-AW was used to form a phosphoenzyme intermediate to which 13H]ouabain binds with high affinity. The data indicate that 1 m~ 2-N3-ATP is nearly as effective as 1 m~ ATP in forming the phosphoenzyme. Control assays demonstrated that the binding of ouabain was abolished for both nucleotides in the absence of nucleotides.

2-Azido-ATP Labeling of Na,K-ATPase 6561

0.ok I I I I I 0.0 0.2 0.4 0.6 0.8 1.0

[2N3-ATP] mM

FIG. 3. Hydrolysis of 2-Ns-ATP by Na,K-ATPase. Purified Na,K- ATPase was diluted to 0.1 mg/ml in 25 m~ imidazole-HC1, pH 7 .4 , l m~ Na2EDTA. 10 pl of the enzyme was mixed with 80 pl of 20 m~ HEPES- triethylamine, pH 7.4, 1.5 m~ phosphoenolpyruvate, 125 m~ NaC1, 25 m~ KCl, 2 n" MgCl,, 0.12 m~ dithiothreitol, 0.315 m g h l p-NADH reduced form, 25 &ml pyruvate kinase, 25 p g / d lactate dehydrogenase. The mixture was placed in a 50-pl microcuvette and was incubated at 37 "C for 1 min. 10 pl of 10-fold concentrated Z-N,-ATP was added. Final concentrations of P-N,-ATP vaned from 0.5 p~ to 1 m ~ . The use of p-NADH was monitored in a Kontron spectrophotometer for about 5 min at 37 "C. The data were plotted using the program Enzfitter", and from the plot a K, of about 50 p~ and a V,, of about 0.88 pmoVmin/mg were extracted. The inset shows the binding of c3H1ouabain to Na,K- ATPase phosphoenzyme formed using Z-N,-ATP as the substrate.

Stoichiometry of Labeling by 2-N3-ATP-The previous results demonstrate that 2-N3-ATP binds to the high-affinity ATP substrate site on Na,K-ATPase and that the enzyme is inacti- vated by 2-N3-ATP in the presence of ultraviolet light. If inactivation of the Na,K-ATPase is caused by photochemical acti- vation of the azide in the active site of the enzyme, a linear relationship between the extent of enzyme inactivation and the extent of incorporation of the probe into the protein is expected. Fig. 4 shows a plot of the ATP-protectable labeling of the a subunit of the Na,K-ATPase versus the percentage of ATP- protectable inhibition of enzymatic activity. A linear relationship up to a maximum inhibition of 50%, corresponding to incorporation of about 0.9 nmol of [a-32P12-N3-ATP, was observed. As discussed above, the nonspecific inactivation of enzyme activity by prolonged irradiation limited the data collec- tion to this range of inhibition. SDS-PAGE and autoradiography (data not shown) demonstrated that all of the [a-32P12-N3-ATP was incorporated into the a subunit of Na,K- ATPase and that the labeling of the a subunit was completely prevented by ATP. Extrapolation of the data shown in Fig. 4 to 100% inactivation indicates that complete inhibition of activity occurs at a stoichiometry of incorporation slightly less than the binding capacity of 2.4 nmol ATP/mg of protein for this enzyme. The difference may be due to some loss of the probe from the protein by hydrolysis during sample handling, as discussed below.

Purification of 2-N3-ATP Labeled Peptides-Partial trypsin digestion of [CX-~'P]~-N~-ATP labeled Na,K-ATPase in the presence of KC1 (23) demonstrated that most of the radiolabel was stably incorporated into the carboxyl-terminal 58 kDa of the a subunit (data not shown). A small amount of radioactivity was also found in the position of the amino-terminal 43-kDa fragment. The p subunit was not labeled. Exhaustive trypsin digestion of the labeled protein was performed using a 1/10 (w/w)

ratio of trypsidATPase, for 4 h at 37 '. After removal of the membranes from the digest by centrifugation, nearly all of the radioactivity was recovered in the supernatant. Greater than 99% of the radioactivity in the supernatant, however, eluted from a Cle HPLC column within 1-2 min aRer injection. Sev- eral HPLC solvent systems were tested in an effort to bind greater amounts of radiolabel to the column; however, none were successful. SDS-PAGE was used to fractionate the digest supernatant to estimate the extent of digestion of the labeled protein and to estimate the sizes of the resultant tryptic peptides. Fig. 5 shows the results of one such experiment in which a 10% polyacrylamide gel was used to fractionate the digest. AS expected, the radiolabel is located exclusively within the (x subunit of Na,K-ATPase before trypsin digestion (lane 1 ). When the digest was separated into supernatant (lane 2 ) and pellet (lane 3 ) fractions, an abundant radiolabeled 30-kDa fragment was seen in the supernatant. The fragment is very resistant to digestion by trypsin, and incubation of the labeled Na,K- ATPase with a 1/10 (w/w) ratio of trypsidATPase for 0.5-4 h at 37 "C did not reduce the yield of the fragment by more than 50%. Approximately 30% of the radioactivity initially incorporated into the a subunit is recovered in the 30-kDa fragment after about 1 h of digestion under these conditions. Incubation of the digest supernatant containing the 30-kDa fragment with 5 M urea for 3 h at 37 "C did not change the mobility of the fragment on SDS-PAGE, indicating that the 30-kDa fragment is a single polypeptide. The 30-kDa fragment was transferred to polyvinylidene difluoride membranes for amino acid sequence determination, and an amino-terminal sequence N-L- E-A-V- was obtained. This sequence corresponds to amino acids 353-357 of the sheep (24) and dog2 Na,K-ATPase a1 subunit. These results indicate that trypsin digestion of [a-32P12-N3- ATP-labeled Na,K-ATPase generates a trypsin-resistant 30-kDa fragment that contains the bound nucleotide. Fig. 5 also shows that a radiolabeled fragment of approximately 17 kDa is also present in the digest supernatant. It is not known whether this fragment is derived from the 30-kDa fragment or not, and this fragment has not yet been isolated in sufficient yield for further analysis.

Trypsin cleavage of the 30-kDa fragment was accomplished after incubation of the initial digest supernatant in 5 M urea, followed by dilution to a final urea concentration of 2.5 M. SDS- PAGE analysis indicated that the 30-kDa fragment was cleaved by trypsin into peptides of about 2.8-15 kDa after urea denaturation. Fig. 6, top panel, shows a typical HPLC chromatogram of peptides generated by this reaction and the location in the chromatogram of the radioactivity associated with the labeled peptides. Two peaks of radioactivity (1 and 2 ) eluting between 54 and 62 min can be seen within the region of peptide fractionation. Both of these peaks of radioactivity are nearly absent in a sample treated identically except that the photochemical labeling reaction was done in the presence of 0.2 m~ ATP (data not shown). Fractions eluting between 54 and 58 min were collected as pool 1, fractions eluting between 58 and 62 min were collected as pool 2, and each pool was fractionated again by HPLC. Three peaks of radioactivity ( la - lc ) eluting at 38-42 min, 4-8 min, and 52-54 min were collected from pool 1 (Fig. 6, middle panel ). The radioactivity eluting between 14 and 24 min in this chromatogram did not bind to the HPLC column and eluted before the start of the solvent gradient at 30 min. Each of the peaks, la-IC, was fractionated further by HPLC. The final separation of peak la is shown in Fig. 6, bottom panel. The single peak of coincident radioactivity and absorption a t both 214 and 272 nm eluting between 37 and 37.5

R. A. Farley, T. A. Brown, G. Scheiner-Bobis, B. Horowitz, and K. Wang, unpublished results, and Z. Xie, personal communication.

6562

1 .oo

0.80

0.60

0.40

0.20

2-Azido-ATP Labeling of Na,K-ATPase

0 5 10 15 20 25 30 35 40 45 50

ATP-protected inhibition (%) FIG. 4. Stoichiometry of covalent photochemical labeling of Na,K-ATPase by 2-Ns-ATP. Samples of purified Na,K-ATPase (0.2 mdml, 5

ml final volume) in 10 m~ Tris-HCI, pH 7.0, 10 mM NaCl, 2.5 mM MgC12, 0.2 m~ Na2EDTA, 0.1 m~ glutathione in reduced form were photolyzed with various concentrations of [a-32P]2-N3-ATP from 1 to 50 p~ in the presence and absence of 0.2 m~ ATP. The specific radioactivity was about

determination of remaining ATPase activity. The remainder of each sample was collected by centrifugation at 50,000 rpm in a Ty 65 rotor for 1 h. 100 pCi/pmol. The conditions of labeling were as described in the legend to Fig. 1. 10 pl from each sample was used in a coupled assay for

The supernatants were discarded, and the pellets were suspended in 25 m~ imidazole-HC1, pH 7.4, 1 m~ Na2EDTA buffer and counted for radioactivity (Cerenkov counting). A known amount of the same batch of [a-32P]2-N3-ATP used in labeling was also counted with the sample to calculate the exact specific activity. ATP protectable inhibition is defined as the difference in activity between samples containing the same concentration of [~Y-~~PI~-N,-ATP in the presence and absence of 0.2 m~ ATP. ATP protectable labeling is defined as the difference in radioactivity incorporated between the same set of samples. The line shown is the best fit to the data, and the intercept on the y axis (0.03 nmol/mg) is indistinguishable within experimental error from zero.

t 94 k D a

e 58 kDa

t 30 kDa

t 17 kDa - t Dye front

1 2 3 4 FIG. 5. Trypsin digestion of Na,K-ATPase labeled with [a-s2p12-

Ns-ATP. Purified Na,K-ATPase was labeled with 50 p~ [a-32P12-N3-ATP as described under “Experimental Procedures” and was digested with L-1-tosylamido-2-phenylethyl chloromethyl ketone-treated trypsin at a ratio of 150 for 3 h a t 37 “C in 25 mM imidazole-HC1, pH 7.4, 1 m~ Na2EDTA. The protein concentration was 2 mg/ml. The sample was centrifuged in a 5 65 rotor a t 50,000 rpm for 1 h. Supernatant and pellet were collected, and samples of about 100 pg from each were analyzed by 10% SDS-PAGE. After the gel was stained and destained, an autoradiogram was prepared. Lane 1 is 100 pg of undigested, labeled enzyme, lane 2 is the supernatant of trypsin-digested, labeled enzyme, lane 3 is the pellet of the sample in lane 2, and lane 4 is the trypsin- digested, labeled enzyme before centrifugation.

min was collected. Peak 2 from the original chromatogram was separated into three peaks eluting between 50 and 52 min (peak 2a), 52 and 54 min (peak Zb), and 58 and 60 min (peak 2c) (Fig. 7, top panel). Fig. 7, bottom panel, shows the final fractionation step of peak 2b. The fractions containing coincident radioactivity and absorption at both 214 and 272 nm eluting between 48.5 and 49.5 min were collected. Samples la , lb , IC, 2a, 2b, and 2c were examined by amino acid sequence analysis; however, sequences were obtained only from samples la , IC, and 2b. No sequences were obtained from the sequencer for samples l b and 2c, and sample 2a was recovered in amounts insufficient for analysis.

Fig. 6, middle panel, peak l a contained two peptides in about a 6:l ratio. The most abundant peptide corresponds to the sequence His496-Ar$06 of the sheep Na,K-ATPase a subunit (24), His-Leu-Leu-Val-Met-Lys-X-Ala-Pro-Glu-Arg. The sev- enth cycle of this sequence, corresponding to GlFo2, was blank, consistent with derivatization of this amino acid by 2-N3-ATP. No radioactivity was recovered from the sequencer at any cycle, but because of the presence of the phosphate groups on the probe it is likely that the labeled amino acid was not eluted from the filter of the sequencer. The other sequence in peak l a corresponds to the sequence Va1606-Lys618 of the sheep Na,K- ATPase a subunit. The yield of this sequence was low and no cycles were blank. Fig. 7, top panel, peak 2b contained only one sequence, corresponding to amino acids H ~ S ~ ~ ~ - A ~ $ ’ O , His-Leu- Leu-Val-Met-Lys-X-Ala-Pro-Glu-Arg-Ile-Leu-Asp-Arg. The sev- enth cycle of the sequence, corresponding to Glfo2, was also blank. The amino acid compositions of the samples corresponding to peaks l a and 2b are shown in Table I. A comparison of these data with the amino acid sequences of the peptides indicates that the sequences obtained represent all of the peptides in each sample.

The labeling of Na,K-ATPase with [ (x -~~P]~-N~-ATP and the subsequent analysis of the labeled peptides were performed three times. The sequences H ~ S ~ ~ - A ~ $ ~ and H ~ S ~ ’ ~ - A ~ $ ~ ’ were found in all experiments, and Gly502 was absent in all of these sequences. The peptide sequence Typ8’-Arg495 in the Na,K-ATPase a subunit was also found in reduced yield together with the sequence H~s~’~-A~$~O in two experiments. No amino acids were missing from the sequence Ty~~*~-Arg4’~, however, in either experiment. The sequence of Fig. 6, middle panel, peak IC described above was I-V-E-I-, corresponding to amino acids 470-473 of the sheep kidney a subunit. This sequence was found only in the experiment described above, and amino acids were identified in all cycles of the sequencer. The

2-Azido-ATP Labeling of Na,K-ATPase

900

' 600 r x - -

300

10 20 30 40 50 60 70 80 80 100 E l ~ t m n time (mln)

I

14 24 34 44 54 64 74 04 94 EIuLion t ime imin)

,

0.05

- -

0.03 a <

0.01

30 32 5 35 37.5 40 42.5 Elution time (mi")

[cr-SZP12-Ns-ATP-labeled Na,K-ATPase. [a-SaP12-N3-ATP-labeled FIG. 6. HPLC separation of urea-treated, trypsin-digested

Na,K-ATPase was digested with trypsin and treated with urea as described under "Experimental Procedures." Top panel, the acetonitrile extraction was loaded to a Vydac C18 column and was fractionated using the following conditions. Solvent A was 20 m~ H3P0,-NaOH, pH 6.5, solvent B was 70% acetonitrile, and a gradient of &loo% B in 90

14 24 34 44 54 84 74 84 Elution lima (min)

I

6563

0.w

0.06 : - 0.04 2

P 0.03

0.02

0.01

0.018

- E

o.o'o ?

0.008

40 s 43 45.5 48 50.5 53 55.5 58 Elution time (min)

FIG. 7. HPLC separation of peak 2 from Fig. 6. Top panel, peak 2 (Fig. 6, top panel) was fractionated on a Vydac C4 column under the same conditions used for peak 1 (Fig. 6, middle panel ). Three peaks of radioactivity (2a, 2b, and 2c) were collected. Bottom panel, radioactive peak 2b (top panel ) was desalted on a Vydac C18 column as described in Fig. 6, bottom panel. The peak of coincident absorbance at 214 nm, at 272 nm, and radioactivity was collected for amino acid analysis and amino acid sequencing.

min at a flow rate of 1.5 mVmin was used. Fractions of 3 ml were collected and counted for radioactivity. Absorbance was monitored with a photodiode array detector (LKB 2140, Pharmacia). In this figure, the upper trace shows the absorbance at 214 nm, and the bar graph shows the radioactivity in each fraction. Middle panel, fractions corresponding to peak 1 (top panel) were collected and fractionated on a Vydac C4 column. Solvent A was 10 m~ H,PO,-KOH, pH 2.55, and solvent B was 70% acetonitrile. A linear gradient of 040% of solvent B in 90 min, at a flow rate of 1.5 ml/min, was used. Fractions of 3 ml were collected. The effluent was monitored for absorbance at 214 nm (solid line) and at 272 nm (broken line). The bar graph shows the radioactivity in each fraction. Bottom panel, radioactive peak la (middle panel) was desalted on a Vydac C18 column. Solvent A was 0.05% trifluoroacetic acid, and

dient was the same as in the middle panel. Fractions of 0.75 ml were solvent B was 0.05% trifluoroacetic acid in 70% acetonitrile. The gra-

collected and counted for 32P radioactivity. The effluent was monitored by absorbance at 214 nm (solid line) and at 272 nm (broken line). The bar graph shows the radioactivity in each fraction. The peak with coincident absorbance at 214 nm, absorbance at 272 nm, and radioactivity was collected for amino acid analysis and amino acid sequencing.

6564 2-Azido-ATP Labeling of Na,K-ATPase

TABLE I Amino acid composition of peptides l a and 26

Amino acid

Peptide l a normalized to Val Expected”

Amount Peptide 2b

Amount normalized to Pm Expectedb

pmol pmol A S X

Glx 33 0.53 76 1.23

0.2 31 1.00 1 35

1

Ser 41 1.13

0.66 1

G ~ Y 41 0 8

0.66 1.2 21 0.26 0

His 40 0.65 1.2 0.68

22 1

’4% 58 0.94 1 0.71

65 1

Thr 39 0.63 2.10

0.2 2

Ala 59 0.95 1.2 30 5 0.16 0

Pro 75 1.21 1.2 31 0.97 1

Tyr 89 1.44 1.00 1

Val 62 0 5 0.16

29 0

Met 9 0.15 1.2 14 0.94 1

Ile 16 0.26 0.3 23 0.45 1

Leu 86 1.39 2 72 0.74 1

Phe 2.32

10 0.16 0 2 3

0.06 LYS 49 0.79 1.2 18 0.58 1

0

1.00 1.3

for 85% of the sample, and one from Va160s-Lys618 accounting for 15% of the sample. a The amounts expected for peptide la are numbers of residues expected if the sample contained two peptides, one from accounting

The amounts expected for peptide 2b are numbers of residues expected for a single peptide from His6m-Ar$*0.

possible significance of the presence of these additional peptides in some of the samples is considered in more detail in the “Discussion.”

DISCUSSION

Several experiments reported here indicate that 2-N3-ATP labels Na,K-ATPase from within the high-affinity ATP binding site of the enzyme. Most significantly, Na,K-ATPase catalyzes the hydrolysis of 2-N3-ATP (Fig. 3). Although the rate of hydrolysis is only about 4% of the rate of hydrolysis of ATP, the use of 2-N3-ATP as a substrate by the enzyme confirms the presence of the analog at the active site. The K, for hydrolysis of 2-N3- ATP was about 50 1.1~; however, under the conditions of the assay approximately half of the azido group exists as a tetra- zole isomer (161, and the true K,,, may be somewhat lower. The hydrolysis of 2-N3-ATP by Na,K-ATPase leads to the formation of a phosphoenzyme intermediate to which ouabain binds with high afflnity (Fig. 3). This indicates that the mechanism of hydrolysis of 2-N3-ATP by Na,K-ATPase is probably the same as the mechanism of enzymatic hydrolysis of ATP. 2-N3-ATP also competitively inhibits the binding of ATP to its high-affinity site on the Na,K-ATPase with a Ki that is close to the K , for hydrolysis (Fig. 2). The inactivation of Na,K-ATPase activity by 2-N3-ATP is proportional to the extent of incorporation of the probe into the a subunit of the enzyme (Fig. 4), and the stoichiometry of incorporation at complete inhibition of activity is estimated to be similar to the binding capacity of the enzyme for ATP.

Purification of tryptic peptides from [a-32P12-N3-ATP-labeled Na,K-ATPase localized the radiolabel to the region between His496 and Arg5l0 of the a subunit. Two labeled peptides were purified, corresponding to the sequences His496-Arg506 and H ~ S ~ ~ ~ - A ~ $ ~ O . The residue at the position of Glfo2 was not identified in either peptide, consistent with the derivatization of this amino acid by 2-N3-ATP. Thus, inactivation of Na,K- ATPase by 2-N3-ATP occurs by reaction of the probe with Glfo2 within the ATP binding site of the protein. The covalent linkage of the nucleotide to Gly502 is likely to be the reason for the incomplete cleavage of the trypsin-sensitive bond after Ar206, resulting in the appearance of the radiolabel in two peptides.

Several observations suggest that 2-N3-ATP may label additional amino acids in the ATP binding site of Na,K-ATPase, but instability of the nucleotide-amino acid linkage may make identification of these amino acids difficult. Glfo2 is located in a carboxyl-terminal 58-kDa tryptic fragment of the a subunit of

Na,K-ATPase. A small amount of radiolabel was detected in a nonoverlapping amino-terminal 43-kDa fragment, suggesting that there may be additional amino acids in this region of the protein that can be labeled by the probe. Attempts to purify a radiolabeled peptide from this region of the polypeptide have not yet been successful due to low yield of radioactivity. In sample IC discussed under “Results,” the sequence 4701-V-E-I- was obtained. The peak of radioactivity from the final HPLC purification of this sample was collected in two consecutive fractions. Although both fractions contained the same sequence, the glutamic acid in cycle 3 was absent in the first fraction. This amino acid was present in the sequence from the second fraction, and this sequence was not found in other per- formances of the experiment. One possible explanation for this observation is that the probe labels G ~ u ~ ~ ~ but that the bond between the probe and the amino acid is very labile. The hydrolysis of the linkage may account for the inconsistent results obtained in different experiments. A similar explanation may account for the appearance of the peptide Va1606-Lys618 in sample la, without evidence of labeling. In another perform- ance of the experiment, the peptide Va1606-Lys618 was separated from the radiolabeled peptide H i ~ ~ ~ ~ - A r g 5 ~ , and again it was found to contain no radioactivity. Although this peptide may have simply been a contaminant in sample la, this sequence corresponds to the peptide in yeast proton transport ATPase that was reported by Davis et al. (26) to be labeled by 2-N3-AMP and 2-N3-ATP. The labeled amino acid also was not identified in that report. Similarly, the sequence ljT481-Are95 was obtained from radiolabeled fractions that also contained the sequence in two of the three experiments reported here. In contrast to the sequence for H~s~’~-A~$~O, for which Glgo2 was always absent, all amino acids in the sequence ljT4s1-Arg.195 were identified in both experiments. The sequence TylAS1-Areg5 overlaps with the sequence in Ca- ATPase that is labeled by TNP-8-N3-ATP and has been suggested to be located within the ATP binding site of that enzyme (27). Although the possibility cannot be excluded that the peptides Va1606-Lys618 and TyF‘l-Areg5 may have been contami- nants of the radiolabeled peptides in some experiments reported here, because of the instability of the linkage between 2-N3-ATP and some amino acids, and the suggestion that these peptides may be located in the ATP binding sites of other P-type ATPases, the possibility that amino acids other than Glfo2 are also labeled by 2-N3-ATP must also be considered.

The location of Gly502 within the ATP binding site of Na,K-

2-Azido-ATP Labeling ofNa,K-ATPase 6565

ATPase is consistent with previous results, which indicate that the region of the a subunit between amino acids 470 and 501 is likely to participate in the interaction between the protein and ATP (2-7, 28). The results of the experiments reported here also indicate that FITC, which has previously been shown to react a t Lys501 (2, 3), is located very close to the ATP binding site in FITC-modified Na,K-ATPase. Although much kinetic evidence is consistent with a location for FITC within the ATP binding site, the protection by ATP against FITC inactivation of enzymatic activity and the competitive inhibition of ATP binding by FITC could occur even if the two molecules do not reside at the same site. A secondary structure model of the ATP binding region of P-type ion pumps (29) places the sequence K-G-A-, corresponding to amino acids 501-503 in Na,K-ATPase, in a sharp turn between a p sheet and an a helix. This turn is a highly conserved structural feature of many ATP binding sites. The presence of the two amino acids labeled by FITC and 2-N3- ATP in this sequence supports the interpretation that the sites occupied by the two inhibitors overlap in the ATP binding site.

In Ca-ATPase, a cross-link between Lys492 and A q f F 8 can be induced by glutaraldehyde reaction at the active site of the protein (30). Because these amino acids are widely separated in the primary structure of the polypeptide, it was suggested that the ATP binding site may be composed of amino acids located on separate domains of the protein and that ATP binds in a cleft between these domains. This model might be used to explain the observation described here that trypsin digestion of 2-N3- ATP-labeled Na,K-ATPase does not cleave the protein into limit tryptic peptides but instead generates a protease-resistant 30- kDa fragment that contains the bound nucleotide. The 30-kDa fragment contains approximately 300 amino acids, beginning a t The fragment includes Asp369 that is phosphorylated by ATP and the amino acids around the middle of the polypeptide that are labeled by ATP site-directed reagents. I t is not known exactly where the fragment ends at its carboxyl termi- nus, but from its size it appears to be within a region of high homology among P-type ATPases and close to the position of AreF8 in the Ca-ATPase. Thus, the 30-kDa fragment may include all, or nearly all, of the amino acids on the protein that interact with nucleotides at the active site. The protease-resistant 30-kDa fragment might be formed by the insertion of 2-N3- ATP into a cleft between regions of the protein that are induced to move closer together when a nucleotide occupies the binding site. When the nucleotide is ATP or azido-ATP, this conformational change may facilitate the enzymatic transfer of phosphate from the nucleotide to Asp369. The fortuitous location of the azido group on 2-N3-ATP appears to lock the protein around

the nucleotide, shielding the trypsin-sensitive amide bonds from the protease. The presence of a large number of lysine and arginine residues in the ATP binding site has previously been noted (30). The 30-kDa fragment appears to be too small to include the amino acids Asp71o and Asp714 that have been identified by CIR-ATP modification (9) and Lys719 that has been identified by p-(fluorosulfony1)benzoyladenosine modification (8) of Na,K-ATPase. Either a peptide containing these amino acids can be cleaved by trypsin when the nucleotide occupies the catalytic site or these amino acids may not be directly located in the nucleotide binding site.

Acknowledgments-We thank Michael Kabalin for providing purified Na,K-ATPase for these experiments, and we thank Lynn Williams of the University of Southern California and Dr. Terence Kirley of the Uni- versity of Cincinnati for amino acid sequence and composition analysis. We are grateful to Dr. Charles McKenna (USC) for NMR analysis of 2-N,-ATP.

REFERENCES 1. Jorgensen, P. L., and Andersen, J. F! (1988) J. Membr. Bioi. 103.95-120 2. Farley, R. A,, Tran. C. M., Carilli, C. T., Hawke, D., and Shively, J. E. (1984) J.

3. Kirley, T. L., Wallick, E. T., and Lane, L. K. (1984) Biochem. Biophys. Res. Bioi. Chem. 259,9532-9535

4. Pedemonte, C. H.. Kirley, T. L., Treuheit, M. J., and Kaplan, J. H. (1992)FEBS

5. Hinz, H. R., and Kirley, T. L. (19901 J. Bid. Chem. 266,10260-10265 6. Xu, R. (1989) Biochemistry 28,5764-5772 7. Tran, C. M., Scheiner-Bobis, G., Schoner, W., and Farley, R. A. (1988) Biophys.

8. Ohta, T., Nagano, K.. and Yoshida, M. (1986) Proc. Natl. Acad. Sci. U. 5'. A. 83,

9. Ovchinnikov, Y. A,, Dzhandzugazyan, K. N., Lutsenko, S. V.. Mustayev, A. A,,

Commun. 126, 767-773

Lett. 314.97-100

J. 63,343

2071-2075

and Modyanov, N. N. (1987) FEBS Lett. 217, 111-116 10. Scheiner-Bobis, G., and Schoner, W. (1985) Eur. J. Biochem. 162, 739-746 11. Jorgensen, P. L. (1974) Biochim. Biophys. Acta 366,36-52 12. Schaeffer, H. J., and Thomas, J. (1958) J. Am. Chem. Soc. So, 373S3742 13. Michelson, A. M. (1964) Biochim. Biophys. Acta 91, 1-13 14. Yoshikawa, M., Kato, T., and Takenishi, T. (1967) Rtrahedron Lett. 60,5065-

15. Tomasz. J. (1978) Nucleic Acid Chem. 3,765-769 16. Czarnecki, J. J. (1984) Biochim. Biophys. Acta 800 ,141 17. Boulay, E, Dalbon. P., and Vignais, P. V. (1985) Biochemistry 24. 73727379 18. Symons, R. H. (1968) Biochim. Biophys. Acta 166, 609-610 19. Melese, T., and Boyer, P. (1985) J. Biol. Chem. 280,1539S15401 20. Fiske, C. H., and Subbarow, Y. (1925) J. Biol. Chem. 66,37&400 21. Laemmli, U. K (1970) Nature 227.680485 22. Rush, J., and Konigsberg, W. H. (1990) J. Bid. Chem. 266.48214827 23. Jorgensen, P. L., and Farley, R. A. (1988) Methods Enzymoi. 166,291301 24. Shull, G. E., Schwartz, A., and Lingrel, J. B. (1985) Nature 316,691495 25. Deleted in proof 26. Davis, C. B., Smith, K. E., Campbell, B. N., and Hammes, G. G. (1990) J. Biol.

27. McIntosh, D. B., Woolley, D. G., and Berman. M. C. (1992) J. Biol. Chem. 267,

28. Wang, K., and Farley, R. A. (1992) J. Bioi. Chem. 267, 3577-3580

30. McIntosh, D. B. (19921 J. Bid . Chem. 267,2232S22335 29. Taylor, W. R., and Green, N. M. (1989) Eur. J. Biochem. 179,241-248

5068

Chem. 266, 1300-1305

5301-5309

photochemical labeling and inhibition of na,k-atpase by 2-azido

Documents