Labeling of lysine 492 with pyridoxal 5'-phosphate in the sarcoplasmic reticulum Ca(2+)-ATPase. Lysine 492 residue is located outside the fluorescein 5-isothiocyanate-binding region in or near the ATP binding site.

Sarcoplasmic reticulum vesicles were treated with 2 mM pyridoxal 5'-phosphate (PLP) at 25 degrees C and pH 7.0 for 6 min and reduced by NaBH4. Both the activity of the Ca(2+)-ATPase and the capacity for high affinity Mg-ATP binding were greatly reduced. Acetyl phosphate hydrolysis or phosphoenzyme formation from Pi was not inhibited. The enzyme was protected by high affinity Mg-ATP binding against the PLP-induced inhibition. A similar protective effect was obtained by Mg-AMP as well as by Mg-ADP. Acetyl phosphate or Pi gave no protection. The PLP-treated vesicles were solubilized in SDS, and the Ca(2+)-ATPase was purified by size exclusion high performance liquid chromatography (HPLC). Mapping the fluorescently labeled peptides in the tryptic digest by reversed phase HPLC revealed a single fluorescent peak, which was protected by Mg-ATP against labeling. Sequencing showed that Lys-492 is the residue labeled with PLP. These results demonstrate that Lys-492 is located in or near the ATP binding site but not in the phosphorylation site or the Pi binding site. When Lys-515 was entirely prelabeled with fluorescein 5-isothiocyanate (FITC), the subsequent labeling of Lys-492 with PLP was not prevented. This finding demonstrates that Lys-492 is located outside the FITC-binding region. It has been widely accepted that FITC occupies the adenosine-binding region within the ATP binding site. In contrast to FITC, Mg-AMP strongly inhibited the labeling of Lys-492 with PLP. These findings lead to the conclusion that Lys-492 is located outside the adenosine-binding region, most probably in or near the region occupied by the alpha-phosphoryl group of Mg-ATP bound to the ATP binding site.

Sarcoplasmic reticulum vesicles were treated with 2 mM pyridoxal 5"phosphate (PLP) at 25 "C and pH 7.0 for 6 min and reduced by NaEiHa. Both the activity of the Ca2+-ATPase and the capacity for high affinity Mg-ATP binding were greatly reduced. Acetyl phosphate hydrolysis or phosphoenzyme formation from Pi was not inhibited. The enzyme was protected by high affinity Mg-ATP binding against the PLP-induced inhibition. A similar protective effect was obtained by Mg-AMP as well as by Mg-ADP. Acetyl phosphate or Pi gave no protection. The PLP-treated vesicles were solubilized in SDS, and the Ca2+-ATPase was purified by size exclusion high performance liquid chromatography (HPLC). Mapping the fluorescently labeled peptides in the tryptic digest by reversed phase HPLC revealed a single fluorescent peak, which was protected by Mg-ATP against labeling. Sequencing showed that Lys-492 is the residue labeled with PLP. These results demonstrate that Lys-492 is located in or near the ATP binding site but not in the phosphorylation site or the Pi binding site. When Lys-515 was entirely prelabeled with fluorescein 5-isothiocyanate (FITC), the subsequent labeling of Lys-492 with PLP was not prevented. This finding demonstrates that Lys-492 is located outside the FITC-binding region. It has been widely accepted that FITC occupies the adenosine-binding region within the ATP binding site. In contrast to FITC, Mg-AMP strongly inhibited the labeling of Lys-492 with PLP. These findings lead to the conclusion that Lys-492 is located outside the adenosine-binding region, most probably in or near the region occupied by the a-phosphoryl group of Mg-ATP bound to the ATP binding site.
The Ca2+-ATPase of skeletal muscle SR' catalyzes active Ca2+ transport coupled to ATP hydrolysis (1,2). The enzyme ~ ~ * This work was supported by a grant-in-aid for Scientific Research from the Ministry of Education, Science and Culture, Japan (to T. K.). The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "aduertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. consists of a single 110-kDa polypeptide chain, of which the whole amino acid sequence has been revealed (3). This enzyme has one high affinity ATP binding sitelpolypeptide chain. In the catalytic cycle, when Ca2+ ions bind to the high affinity Ca2+ binding sites and as a result activate the enzyme, the yphosphoryl group of Mg-ATP bound to the ATP binding site is transferred to Asp-351 (3)(4)(5)(6) to form an EP intermediate (7,8). Acetyl phosphate can also serve as a substrate through EP formation (9,lO). In addition, the EP can be formed from Pi by reversal of the catalytic cycle (11, 12).
Affinity labeling has been carried out to identify some residues that may be part of the ATP binding site of the Ca2+-ATPase. Lys-515 is labeled by FITC in competition with ATP (13,141. Adenosine triphosphopyridoxal labels Lys-684 in the presence of Ca2+ (15) and both Lys-684 and Lys-492 in the absence of Ca2+ (16). Lys-492 is also labeled by 2',3'-0-(2,4,6trinitrophenyl)-8-azido-AMP and -ATP (17) and 7-amino-4methylcoumarin-3-acetic acid succinimidyl ester (18). It has been shown by Murphy (19) that incubation of the SR vesicles with PLP followed by NaBH4 reduction results in a loss of the Ca2+-ATPase activity and that the enzyme is protected by ATP against the inhibition. PLP has also been used to label the ATP binding sites in other P-type iontransporting ATPases. PLP inhibits the pig gastric H+,K+-ATPase (20) and lamb kidney Na+,K+-ATPase (21) by labeling Lys-497 in the former and Lys-480 in the latter. The sequences surrounding the sites labeled by PLP are highly conserved in these ATPases. The corresponding lysyl residue is also conserved in the SR Ca*+-ATPase (3).
In this study, we have identified PLP-labeled residues in or near the ATP binding site of the SR Ca2+-ATPase. Further, in an attempt to reveal the topography of the ATP binding site, we have examined the relative spatial location of the bound PLP toward the region occupied by FITC bound to Lys-515. The results show that Lys-492 is located outside the FITC-binding region, most probably in or near the region occupied by the a-phosphoryl group of Mg-ATP bound to the ATP binding site.
Preparation of SR Vesicles-SR vesicles were prepared from rabbit skeletal muscle as described previously (22) with slight modifications and stored in 0.1 mM CaCl2, 0.1 M KC1, 0.3 M sucrose, and 5 mM MOPS/Tris (pH 7.0) at -80 "C.
Labeling with PLP-Labeling was started at 25 "C by adding PLP to a suspension of the SR vesicles in the dark. The mixture had a final composition of 2 mg of SR vesicles/ml, 2 mM PLP, 5 mM MgCl,, 1 mM EGTA, 0.1 M KCI, 0.1 M sucrose, and 50 mM MOPS/NaOH (pH 7.0), unless otherwise stated. The reaction was stopped by adding an equal volume of 6 mM NaBH4, and then the mixture was centrifuged. The pellet was washed twice by centrifugation with a solution (Solution A)  Double Labeling with FZTC and PLP-After the SR vesicles were labeled with FITC, the vesicles were washed twice with Solution A by centrifugation and then suspended in Solution A. The vesicles were then labeled with PLP, as described above unless otherwise stated. The reaction was quenched with NaBH4, and the mixture was centrifuged. The pellet was washed by centrifugation with a solution containing 0.1 mM CaCl,, 50 mM NaCl, 0.3 M sucrose, and 5 mM MOPS/Tris (pH 7.0).
Purification of the FITC-and/or PLP-labeled Ca2+-ATPase by Size Exclusion HPLC and Determination of the Content of Bound PLP-The FITC-and/or PLP-labeled SR vesicles were solubilized in a medium (the elution buffer in the size exclusion HPLC as described below) containing 2% SDS, 0.1 mM CaCl,, 5 mM MgCl,, 100 mM Li2S0,, and 20 mM sodium phosphate (pH 7.0). After insoluble residues were removed by centrifugation, the supernatant was subjected to size exclusion HPLC at room temperature by use of a TSK SWm guard column (0.6 X 4 cm, Tosoh, Japan) and a TSK,, G3000SWxL column (0.78 X 30 cm, Tosoh, Japan). The composition of various fractions was analyzed by SDS gel electrophoresis (data not shown). The fractions almost exclusively composed of 110-kDa protein were collected. The content of bound PLP in the enzyme thus purified was determined from the absorbance at 325 nm with an extinction coefficient of 10,150 (24).
Proteolysis, Peptide Isolation, and Sequencing-The purified Ca2+-ATPase (0.4 mg) collected as above was precipitated by addition of -20 "C acetone, washed with acetone at 4 "C by centrifugation, and dried in a stream of nitrogen. After addition of 0.2 ml of 8 M urea, the pellet was sonicated on a Branson Sonifier model 200 (Danbury, CT) on ice. The enzyme was then digested with TPCK-trypsin (120 of trypsin to ATPase by weight) at 37 'C for 2 h in 0.8 ml of a medium containing 1 mM CaCl, and 20 mM Tris/HCl (pH 8.0). After the second addition of TPCK-trypsin (20 pl, 1:20 of trypsin to ATPase by weight), the enzyme was further digested at 37 "C for 16 h. Insoluble residues were removed by centrifugation, and the clear supernatant was applied to reversed phase HPLC, which was carried out by use of a prepacked reversed phase C,/C,, column SuperPac Pep-S (5 pm, 0.4 X 25 cm, Pharmacia LKB Biotechnology) connected with a UV monitor SPD-1OAV (Shimadzu, Japan) and a fluorescence monitor F-1000 (Hitachi, Japan). The elution was performed at a flow rate of 1 ml/min by use of a gradient pump LC-1OAD (Shimadzu, Japan). The absorbance of peptides was monitored at 214 nm, and the fluorescence of PLP was monitored with excitation at 325 nm and emission at 400 nm. The emitted light was passed through a filter L-40 (Hoya, Japan), which cut off the light below 370 nm. PLPlabeled peptides isolated were sequenced using a protein sequencer 477A (Applied Biosystems, Inc.) connected with a phenylthiohydantoin-derivative analyzer 120A (Applied Biosystems, Inc.).
Ca*+-ATPase Actiuity-The total ATPase activity was determined at 25 "C in a mixture containing 20 pg of SR vesicles/ml, 0. The CaZ+-ATPase activity was obtained by subtracting the CaZ+independent ATPase activity (determined in the presence of 5 mM EGTA without added CaC1,) from the total ATPase activity.
Acetyl Phosphute Hydrolysis-Hydrolysis of acetyl phosphate was performed at 25 "C in a mixture containing 0.1 mg of SR vesicles/ml, 1 mM acetyl phosphate, 5 mM MgCl,, 0.5 mM CaCl,, 0.4 mM EGTA, 2 p~ A23187, 0.1 M KCl, and 50 mM MOPS/Tris (pH 7.0). The reaction was quenched by addition of a neutralized hydroxylamine solution, and the amount of acetyl phosphate remaining was determined by the method of Lipmann and Tuttle (25). The Ca2+-independent hydrolysis of acetyl phosphate (determined in the presence of 5 mM EGTA without added CaC12) was negligible.
Phosphorylation of the Ca2+-ATPase with ATP or P;-Phosphorylation with [y-32P]ATP was performed at 25 "C for 5 s in a mixture containing 0. 3  ATP, and Mg-ADP in the absence of added AMP were calculated from absolute stability constants listed by Fabiato and Fabiato (29) by use of a program according to the algorithm of Fabiato and Fabiato (29) or that of Newton and Raphson, taking into account pH, temperature, and ionic strength. Concentrations of Mg2+ and Mg-AMP in the presence of added AMP were calculated from the absolute stability constants listed by Sillen and Martell (30). Protein concentrations were determined by the method of Lowry et al. (31) with bovine serum albumin as a standard.

RESULTS
Inhibition of the Ca2'-ATPase by PLP-The SR vesicles were treated with 2 mM PLP for various times, and the reaction was quenched with NaBH,. The Ca2+-ATPase activity decreased progressively and reached a steady level (20% of the control level) in 40-60 min (Fig. 1, Table I). This inhibition is irreversible because the activity was not restored by washing the vesicles.
Protection of the Ca2+-ATPase by Various Ligands against PLP-induced Inhibition-When the vesicles were treated with 2 mM PLP in the presence of Mg-ATP for 6 min, the Ca2+-ATPase was strongly protected by the Mg-ATP against the PLP-induced inhibition (Fig, 2A) The SR vesicles were treated with PLP as described under "Experimental Procedures" (0). For the control, the treatment was performed without PLP, otherwise as above (0). At the times indicated, the reaction was quenched with NaBH,; and the Ca2+-ATPase activity was determined.

TABLE I Protection of the Ca2+-ATPase by various ligands against PLPinduced inhibition
In the experiment where no ligand was added, the SR vesicles were treated with PLP as described for Fig. 1. In the experiments with added ligands, the concentration of M e was adjusted to 5 mM by addition of various concentrations of MgCl,. In the experiment with added Mg-AMP, the ionic strength was adjusted to 0.58 by addition of tetraethylammonium chloride. At 6 or 60 min after the start of the treatment, the reaction was quenched with NaBH, and the Ca2+-ATPase activity was determined. The inhibition of the Ca2+-ATPase by PLP was unaffected by the presence of tetraethylammonium chloride. In the control, the Ca2+-ATPase activity remained constant during the 60-min incubation without PLP in the absence and presence of tetraethylammonium chloride. treated with 2 mM PLP for 60 min, the enzyme was again protected by Mg-ATP (Fig. 2B). The concentration of Mg-ATP giving a half-maximum protection was 2.1 mM. The inhibition remaining to a small extent even at high concentrations of Mg-ATP is likely due to binding of PLP to the unidentified site(s) other than the ATP binding site.
The Ca2+-ATPase was also strongly protected by either Mg-ADP or Mg-AMP (Table I). However, high concentrations of Mg-AMP were required for this protection. This is due to the low affinity of this enzyme for AMP in agreement with that reported previously (32). In contrast, acetyl phosphate or Pi gave no protection.
Lack of ATP Binding to the Ca2+-ATPase of PLP-treated SR Vesicles-In the control in which the SR vesicles were treated without PLP for 6 min, the maximum level of ATP binding to the vesicles and the dissociation constant for Mg-ATP were 3.9 nmol/mg and 4.4 PM, respectively (Fig. 3). This maximum level of ATP binding is in good agreement with the content of the phosphorylation site in the SR vesicles used affinity for Mg-ATP is in accord with that of the intact SR Ca2'-ATPase reported previously (33). When the vesicles were treated with PLP for 6 min, the extent of ATP binding was reduced to 13% of that in the control. These results show that Mg-ATP cannot bind to the PLP-labeled Ca2+-ATPase.
Effects of Labeling with PLP on EP Formation and Acetyl Phosphate Hydrolysis-EP formation from ATP was strongly inhibited by the 6-min treatment with PLP (Table 11). This is consistent with the findings that the treatment caused a pronounced inhibition of the Ca2+-ATPase and Mg-ATP binding (Figs. 1 Table I). In sharp contrast, EP formation from Pi was not inhibited by the treatment with PLP for 6 or 60 min. The acetyl phosphatase activity was reduced only slightly by the 6-min treatment with PLP, while it was reduced to an appreciable extent by the 60-min treatment.

and 3, and
Binding of PLP to FITC-prelabeled or Unprelabeled SR Vesicles in the Absence and Presence of Mg-ATP-The content of bound PLP in the Ca2+-ATPase purified from the PLP-labeled SR vesicles was determined from the 325-nm absorbance of the pyridoxal moiety (Fig. 4A, inset). When the SR vesicles, which were not prelabeled with FITC, were treated with PLP in the absence of Mg-ATP, PLP binding occurred in a biphasic manner (Fig. 4A, main figure). The binding increased rapidly during the initial 6-min incubation and further proceeded very slowly up to 60 min. When the vesicles were treated with PLP in the presence of 20 mM Mg-ATP, the initial rapid binding was markedly depressed, and the subsequent slow phase of the binding disappeared completely.
When the vesicles were prelabeled with FITC, the rate and extent of the PLP binding were reduced appreciably. Nevertheless, the extent of the PLP binding with the FITC-prelabeled vesicles was significantly larger than that of the PLP binding in the presence of 20 mM Mg-ATP. When the vesicles were treated with PLP for 60 min, the Mg-ATP-sensitive part of the content of bound PLP in the vesicles prelabeled with FITC was approximately half that of the content of bound PLP in the vesicles that were not prelabeled with FITC.
A plot of the Ca2+-ATPase activity versus the Mg-ATPsensitive part of the content of bound PLP showed a concave curve (Fig. 4B). Extrapolation to 0% residual ATPase activity gave 10-11 nmol of bound PLP/mg of the purified ATPase protein. This value is about twice the content of the phosphorylation site (5.1 nmol/mg of the purified ATPase protein), which was determined by purification of the phosphorylated Ca2+-ATPase as described under "Experimental Procedures." This is consistent with the findings that two distinct ATP-sensitive sites were labeled with PLP by the 60-min treatment as described later ( Fig. 6A and B, and Table 111).
Evaluation by HPLC Peptide Mapping of Effects of Mg-ATP, Mg-ADP, Mg-AMP, and FITC on the 6-min PLP Labeling-The SR vesicles were treated with PLP for 6 min, and the tryptic digest of the Ca2+-ATPase purified from the vesi- The SR vesicles were treated with PLP for 6 or 60 min, otherwise as described in legend to Fig. 1. For the control, the vesicles were treated without PLP for 6 min, otherwise as above. After the reaction was quenched with NaBH,, EP formation from ATP or Pi as well as acetyl phosphate hydrolysis was determined. The capacity for EP formation from ATP or Pi and the acetyl phosphatase activity remained constant during the 60-min treatment without PLP. cles was subjected to reversed phase HPLC (Fig. 5A). When 5 mM Mg-ATP was included in the reaction, only one ( Fig.  5A, peak 1 ) of the major fluorescent peaks disappeared completely (Fig. 5B). When 3.8 mM Mg-ADP was included, peak 1 again disappeared completely (data not shown). When 110 mM Mg-AMP was included, peak 1 was greatly reduced (compare Fig. 5, E with D). In fact, the area of peak 1 was only 15% of that of peak 1 in the control given in Fig. 5 0 . In contrast, peak 1 was reduced only to a limited extent by the pretreatment with FITC (compare Fig. 5, C with A). The area of this peak 1 was actually 47% that ofpeak 1 in Fig. 5A. This is consistent with the finding (Fig. 4A) that the Mg-ATPsensitive part of the content of bound PLP in the vesicles treated with PLP for 6 min was reduced to 45% by the pretreatment with FITC. The results given in Fig. 5 were unaffected by the presence of added 100 NM Caz+ (data not shown). In D and E, the ionic strength was adjusted to 0.58 with tetraethylammonium chloride. Otherwise, the conditions for the PLP labeling were as described under "Experimental Procedures." The Ca2+-ATPase was purified, digested with TPCK-trypsin, and subjected to reversed phase HPLC as described under "Experimental Procedures." The elution was performed with the following linear gradient of acetonitrile in 0.1% trifluoroacetic acid; 0% from 0 to 10 min, 10% at 20 min, 25% at 80 min, and 48% at 120 min. The absorbance (Absorb.) of peptides (upper trace) and fluorescence (Fluores.) of PLP (lower trace) were monitored.
Evaluation by HPLC Peptide Mapping of Effects of Mg-ATP, Mg-ADP, Mg-AMP, and FITC on the 60-min PLP Labeling-The tryptic digest of the Ca2+-ATPase purified from the SR vesicles, which were treated with PLP for 60 min, was subjected to reversed phase HPLC (Fig. 6A). The 60-min treatment yielded an additional fluorescent peak (peak 2); otherwise, the fluorochromatogram was the same as that obtained by the 6-min treatment (cf. Fig. 5A). The area of peak 2 was virtually equal to that of peak 1. When 20 mM Mg-ATP was included in the reaction, both peak 1 and peak 2 disappeared completely (Fig. 6B). When 16 mM Mg-ADP was included, both the peaks again disappeared (data not shown). When 110 mM Mg-AMP was included, peak 1 and peak 2 were greatly reduced (compare Fig. 6, E with D); the areas of peak 1 and peak 2 were 26 and 23%, respectively, those of peak 1 and peak 2 in the control given in Fig. 6D. When the vesicles were pretreated with FITC, peak 1 was reduced only slightly (compare Fig. 6, C with A); the area of peak 1 was 79% that of peak 1 in Fig. 6A. In contrast, peak 2 was completely abolished. Thus, the sum of the areas of peak 1 andpeak 2 was reduced to 40%. These results are consistent with the finding (Fig. 4A) that the Mg-ATP-sensitive part of the content of bound PLP in the vesicles treated with PLP for 60 min was reduced to 41% by the pretreatment with FITC. The results given in Fig. 6 were again unaffected by the presence of added 100 MM Ca2+ (data not shown).
Amino Acid Sequence Analysis of the PLP-labeled Peptides-The peptides within peak 1 and peak 2 were purified repeatedly by reversed phase HPLC. Each peak contained only one unique PLP-labeled peptide. The PLP-labeled peptides isolated were sequenced (Table 111)  Amino acid sequence anulysis of the PLP-labeled peptides Fractions containing peak I and peak 2 in Fig. 6A were pooled separately. Peptide 1 from peak 1 and Peptide 2 from peak 2 were purified in three steps by reversed phase HPLC using the same column as described for Fig. 6. In step 1, a linear gradient of acetonitrile in 5 mM sodium phosphate (pH 6.9) and 20 mM Na2S04 was used. In step 2, a linear gradient of acetonitrile in 0.1% ammonium trifluoroacetate (pH 6.5) was used. In step 3, the sample containing Peptide I or Peptide 2 was desalted using a linear gradient of acetonitrile in 0.1% trifluoroacetic acid. The peptides thus isolated were sequenced by the protein sequenator. The dash (-) indicates that no phenylthiohydantoin-derivatives were detected.  515 could be detected at the corresponding cycles. This suggests that these residues were labeled with PLP.

DISCUSSION
The present results show that labeling of Lys-492 with no appreciable labeling of Lys-515 in the 6-min treatment with PLP ( Fig. 5A) is accompanied by a strong inhibition of the Ca2+-ATPase ( Fig. 1 and Table I) and that an exclusive prevention of the labeling of Lys-492 ( Fig. 5B and E) is accompanied by a pronounced protection of the enzyme against the inhibition ( Fig. 2A and Table I). These findings demonstrate that this enzyme inhibition is due to labeling of Lys-492 with PLP. Accordingly, the observed protection of the enzyme by high affinity binding of Mg-ATP ( Fig. 2A) gives evidence that Lys-492 is located in or near the high affinity ATP binding site. This conclusion is further supported by the findings that the high affinity binding of Mg-ATP is nearly completely prevented by the 6-min treatment with PLP (Fig. 3) and that the labeling of Lys-492 with PLP is completely prevented by the presence of Mg-ATP (Figs. 5B and 6B).
In some experiments, Lys-515 of the Ca2+-ATPase has been exclusively prelabeled with FITC (see "Experimental Procedures''). The content of bound FITC, being 9.4 nmol/mg of the purified ATPase protein, indicates that almost all the Lys-515 residues of the 110-kDa Ca2+-ATPase present in the SR vesicles have been prelabeled with FITC. This conclusion is consistent with the finding (Fig. 6C)  It has been widely accepted from x-ray crystallographic studies (36,37) that the fluorescein moieties of fluorescein derivatives bound to the nucleotide-binding sites in a variety of enzymes fill the adenosine-binding regions within the sites.
Most probably, this is also valid in the case of the SR Ca2+-ATPase because it has been shown that the labeling with FITC prevents high affinity ATP binding to this enzyme (13, 38) but inhibits neither acetyl phosphate-supported enzyme activity nor EP formation from Pi (13,39,40). Thus, the observed lack of a prevention of PLP labeling of Lys-492 by the FITC prelabeling leads to the conclusion that Lys-492 is located outside the adenosine-binding region of the ATP binding site.
The present results further demonstrate that Lys-492 is located neither in the phosphorylation site nor in the Pi binding site because the labeling of Lys-492 with PLP does not inhibit acetyl phosphate hydrolysis or EP formation from Pi (Table 11) and because the Ca2+-ATPase is not protected by acetyl phosphate or Pi against the PLP-induced inhibition ( Table I). This lack of inhibition of acetyl phosphate hydrolysis or EP formation from Pi is consistent with the previously reported observations (17) that phosphorylation of the Ca2+-ATPase with acetyl phosphate or Pi is not prevented by labeling of Lys-492 with 2',3'-0-(2,4,6-trinitrophenyl)-8azido-AMP. In addition, in the present experiments Mg-AMP has greatly inhibited both the labeling of Lys-492 with PLP ( Fig. 5E) and the PLP-induced enzyme inhibition (Table I).
Taken together, these findings strongly suggest that Lys-492 is located in or near the region occupied by the a-phosphoryl group of Mg-ATP bound to the ATP binding site.
It appears possible that the electrostatic repulsion between the negatively charged PLP bound to Lys-492 and the aphosphoryl group of Mg-ATP is involved in the observed inhibition of Mg-ATP binding (Fig. 3). This possibility is in harmony with the previously reported findings that the affinity of the Na+,K+-ATPase for Mg-ATP is reduced by substitution of an acidic amino acid for the corresponding Lys-480 conserved in this enzyme (41) and that the Ca*+-ATPase is not inhibited by labeling of Lys-492 with 7-amino-4-methylcoumarin-3-acetic acid succinimidyl ester, which has no negative charge (18).
When Lys-492 has been labeled with PLP, the extent of the inhibition of Mg-ATP binding (87% inhibition, Fig. 3) is significantly larger than that of the inhibition of the Ca2+-ATPase (64% inhibition, Table I). This discrepancy remains unsolved, but it might be due to a possible Ca2+-induced change in the relative spatial location of the bound PLP to the a-phosphoryl group of Mg-ATP within the ATP binding site (Mg-ATP binding has been determined in the absence of Ca2+, while the ATPase activity was determined in the presence of Ca2+). However, contrary to this possibility, no Ca2+induced change in the reactivity of the surrounding residues toward PLP has been found in the present experiments (see text under result^'^). This finding is in harmony with the recent observations (18) that the efficiency of fluorescence energy transfer between 7-amino-4-methylcoumarin-3-acetic acid succinimidyl ester and FITC labels on the Ca2+-ATPase is unaffected by high affinity Ca2+ binding, but contrasts with the previously reported findings (15, 16) that the target site specificity of adenosine triphosphopyridoxal changes on bind-Labeling of SR Ca2+-ATPase ing of Ca" to the Ca*+-ATPase.