Modification of Arginine-198 in Sarcoplasmic Reticulum Ca2+-ATPase by 1,2-Cyclohexanedione Causes Inhibition of Formation of the Phosphoenzyme Intermediate from Inorganic Phosphate*

Sarcoplasmic reticulum vesicles were modified with 1,2-cyclohexanedione (CHD), a specific arginine-modifying reagent, in sodium borate (pH 8.0 or 8.8). Phosphoenzyme formation from Pi in the Ca2+-ATPase (reversal of hydrolysis of the phosphoenzyme intermediate) was almost completely inhibited by the modification with CHD. Tight binding of F− and Mg2+ and high affinity binding of vanadate in the presence of Mg2+, either of which produces a transition state analog for phosphoenzyme formation from the magnesium-enzyme-phosphate complex, were also markedly inhibited. In contrast, phosphoenzyme formation from acetyl phosphate in the forward reaction was unaffected. The enzyme was appreciably protected by tight binding of F− and Mg2+ or by high affinity binding of vanadate in the presence of Mg2+, but not by the presence of 20 mm MgCl2 alone or 150 mm Pi alone, against the CHD-induced inhibition of phosphoenzyme formation from Pi. Peptide mapping of the tryptic digests, detection of peptides containing CHD-modified arginyl residues with Girard’s reagent T, sequencing, and mass spectrometry showed that Arg-198 was a single major residue protected by tight binding of F− and Mg2+ against the modification with CHD. These results indicate that modification of Arg-198 with CHD is responsible for at least a part (the portion reduced by the transition state analogs) of the CHD-induced inhibition of phosphoenzyme formation from Pi and suggest that Arg-198 is located in or close to the catalytic site in the transition state for phosphoenzyme formation from the magnesium-enzyme-phosphate complex.

. Finally, the EP is hydrolyzed to form P i and the dephosphoenzyme. Acetyl phosphate also serves as a substrate through EP formation (10,11). The EP can be formed from P i in the presence of Mg 2ϩ and absence of Ca 2ϩ by reversal of the late step of the catalytic cycle (12,13). This EP formation occurs through a magnesium-enzyme-phosphate complex that is formed by random binding of Mg 2ϩ and P i to the enzyme (14,15).
CHD is a chemical modification reagent that reacts specifically with the guanidino group of arginyl residues to produce a stable product, DHCH-Arg (16). Recently, we have shown that binding of Mg⅐ATP or Mg⅐ADP to the ATP binding site of the Ca 2ϩ -ATPase is completely inhibited by the modification with CHD and that Arg-489 and Arg-678 are involved in this inhibition (17). The findings have led to the conclusion that these arginyl residues are located at the ATP binding site. We have further shown that EP formation from P i has also been inhibited by the modification with CHD and suggested that the arginyl residue(s) involved in this inhibition is distinct from the above residues.
It was previously shown (18 -20) that F Ϫ and Mg 2ϩ bind simultaneously and tightly to the catalytic site of this enzyme to form a stable transition state analog for EP formation from the magnesium-enzyme-phosphate complex. Vanadate also binds with high affinity to the enzyme in the presence of Mg 2ϩ to form a transition state analog for this EP formation (21,22).
In the present study, to identify the arginyl residue(s) involved in the CHD-induced inhibition of EP formation from P i , we have modified the SR Ca 2ϩ -ATPase with CHD and examined effects of the above transition state analogs on this modification. We have found that the enzyme has been appreciably protected by the transition state analogs against the CHDinduced inhibition of EP formation from P i . Peptide mapping of tryptic digests of the CHD-modified enzyme, sequencing, and mass spectrometry have shown that Arg-198 is a single major residue protected by tight binding of F Ϫ and Mg 2ϩ against the modification with CHD. The results indicate that modification of Arg-198 with CHD is responsible for at least a part (the portion reduced by the transition state analogs) of the inhibition of EP formation from P i and suggest that Arg-198 is located in or close to the catalytic site in the transition state for EP formation from the magnesium-enzyme-phosphate complex.

EXPERIMENTAL PROCEDURES
Preparation of SR Vesicles-SR vesicles were prepared from rabbit skeletal muscle and stored at Ϫ80°C as described previously (23). The content of phosphorylation site determined with [␥-32 P]ATP according to Barrabin et al. (24) was 4.00 Ϯ 0.06 nmol/mg (n ϭ 6).
Pretreatment of SR Vesicles with F Ϫ and Mg 2ϩ -Pretreatment of the SR vesicles with F Ϫ and Mg 2ϩ was performed as described previously (20) with slight modifications. The vesicles (2 mg/ml) were incubated at 25°C for 3 h in 1 mM KF, 10 mM MgCl 2 , 1 mM EGTA, 20% (v/v) Me 2 SO, 20% (v/v) glycerol, 0.1 M KCl, and 40 mM imidazole/HCl (pH 7.5), unless otherwise stated. The reaction was quenched by diluting the mixture twice with an ice-cold solution containing 0.1 mM CaCl 2 , 0.1 M KCl, 0.3 M sucrose, and 5 mM MOPS/Tris (pH 7.0). The resulting vesicles were washed by centrifugation once with this solution.
Modification with CHD-Modification was started at 37°C by adding CHD (dissolved in 50 mM sodium borate (pH 8.0 or 8.8)) to a suspension of the SR vesicles to give a final composition as described in the figure legends. The modification was quenched at 0°C by one of the following methods. In Method I, the mixture was applied to a centrifuge column (10-ml disposable syringe filled with coarse Sephadex G-50) preequilibrated with a solution (Solution A) containing 0.1 mM CaCl 2 , 0.1 M KCl, 20 mM MOPS, and 5 mM sodium borate (pH 7.0). In Method II, the mixture was centrifuged, and the pellet was washed with Solution A and suspended in this solution. In Method III, the mixture was centrifuged, and the pellet was washed with 50 mM sodium borate (pH 8.0) and suspended in this buffer.
Determination of Tightly Bound F Ϫ and Mg 2ϩ -The SR vesicles were treated with F Ϫ and Mg 2ϩ as described above. In the control samples, CaCl 2 was added to the incubation medium to give 0.1 mM free Ca 2ϩ under the otherwise same conditions as above (tight binding of F Ϫ and Mg 2ϩ is prevented by 0.1 mM Ca 2ϩ (20)). The treated vesicles were washed by centrifugation four times with a solution containing 1 M A23187, 2 mM EDTA, 10% (v/v) Me 2 SO, 0.1 M KCl, and 5 mM MOPS/ NaOH (pH 7.0) and suspended in deionized water. Magnesium bound to the vesicles was extracted with 0.8 N HNO 3 , and the concentration of magnesium in the extract was determined by atomic absorption spectrophotometry as described previously (19). The content of tightly bound Mg 2ϩ was obtained by subtracting the content of magnesium in the extract from the control sample. Fluoride bound to the vesicles was extracted by incubating the vesicles at 95°C for 5 min in 5 mM HEPES/ KOH (pH 8.0). The sample was then centrifuged to remove insoluble materials, and KNO 3 was added to the supernatant to give 0.1 M. The sample was adjusted to pH 3.0 with citric acid, and the concentration of F Ϫ was measured by use of a fluoride-selective electrode as described FIG. 1. CHD-induced inhibition of EP formation from P i and protection by pretreatment with F ؊ and Mg 2؉ , by the presence of vanadate and Mg 2؉ , or by EP formation from P i against the inhibition. A, the SR vesicles were pretreated with KF and MgCl 2 (E, q), without KF and with MgCl 2 (Ⅺ, f), with KF and without MgCl 2 in the presence of 5 mM EDTA and absence of EGTA (Ç, å), or with neither KF nor MgCl 2 in the presence of 5 mM EDTA and absence of EGTA (É, ç), otherwise as described under "Experimental Procedures." The vesicles (2 mg/ml) were then incubated with (E, Ⅺ, Ç, É) or without (q, f, å, ç) 4 mM CHD for various times in 2 mM EDTA, 20% (v/v) Me 2 SO, and 30 mM sodium borate (pH 8.8). The reaction was quenched by Method I. The vesicles were further incubated at 25°C for 60 min in 20 mM CaCl 2 , 0.1 M KCl, 20 mM MOPS, and 5 mM sodium borate (pH 7.0) (tightly bound F Ϫ and Mg 2ϩ are entirely released by incubation with 20 mM Ca 2ϩ (19)) and then passed through a centrifuge column preequili-  (45 and 46)). The vesicles were passed through two successive centrifuge columns, the first column being preequilibrated with Solution A and the second column with Solution B. EP formation from 32 P i was determined. Solid and dashed lines show least squares fit to a single exponential in which the first order rate constants were 0.023 and 0.011 min Ϫ1 , respectively. C, the SR vesicles (2.1 mg/ml) were preincubated in the presence of 1.04 mM P i and 8.33 mM MgCl 2 at 37°C for 15 min in 2.08 mM EGTA, 31.3% (v/v) Me 2 SO, and 31.3 mM sodium borate (pH 8.0) (E, q). The amount of EP formed from P i by this preincubation was 4.1 nmol/mg. In other experiments, the vesicles were preincubated in the absence of P i and presence of 8.33 mM MgCl 2 (Ⅺ, f), in the presence of 156 mM P i and 2.1 mM EDTA and absence of MgCl 2 and EGTA (Ç, å), or in the absence of P i , MgCl 2 , and EGTA and presence of 2.1 mM EDTA (É, ç), as above (no EP is formed under these conditions). After the addition of sodium borate (pH 8.0) with (E, Ⅺ, Ç, É) or without (q, f, å, ç) CHD, the vesicles were further incubated for various times. The final composition was 2 mg of the vesicles/ml, 0 or 15 mM CHD, 0, 1, or 150 mM P i , 0 or 8 mM MgCl 2 , 0 or 2 mM EDTA, 0 or 2 mM EGTA, 30% (v/v) Me 2 SO, and 30 mM sodium borate (pH 8.0). The reaction was quenched by Method II. The vesicles were then passed through a centrifuge column preequilibrated with Solution B. EP formation from 32 P i was determined. Solid and dashed lines show least squares fit to a single exponential, in which the first order rate constants were 0.031 and 0.018 min Ϫ1 , respectively.
previously (19). The content of tightly bound F Ϫ was obtained by subtracting the content of F Ϫ in the extract from the control sample.
Determination of Bound Vanadate-The SR vesicles (0.2 mg/ml) were incubated at 25°C for 30 min in various concentrations of vanadate, 40% (v/v) Me 2 SO, 0.1 M KCl, 30 mM MOPS, 7 mM sodium borate (pH 7.0), and others as described in the legend to Fig. 3. The mixture was centrifuged, and the pellet was dissolved in 2% (w/v) SDS and 0.1 N NaOH. The solution was neutralized with HCl, and the concentration of vanadate was measured by the method of Goodno (25).
Proteolysis, Peptide Mapping, Detection of Peptides Containing DHCH-Arg, Sequencing, and Mass Spectrometry-The CHD-modified SR vesicles (1 mg/ml) were digested with TPCK-treated trypsin (0.2 mg/ml) at 40°C for 4 h in 10 mM CaCl 2 and 50 mM sodium borate (pH 8.0). After centrifugation, the supernatant was subjected to reversed phase HPLC that was performed at a flow rate of 1 ml/min as described previously (26). The absorbance of peptides was monitored at 214 nm, and fractions of 0.3-1.4 ml each were collected. Peptides containing DHCH-Arg in the fractions were detected at 325 nm by the method of Patthy et al. (27) using Girard's reagent T. It was difficult to determine the content of DHCH-Arg in the CHD-modified vesicles by this method because the background level of the absorbance was too high even after solubilization of the vesicles with SDS. Sequencing of isolated peptides was performed with an Applied Biosystems 477A/120A sequencer. For mass spectrometry, the isolated peptides were dissolved in 0.1% trifluoroacetic acid and 0.3% 1-thioglycerol and infused into the frit-FAB (fast atom bombardment) probe of a Jeol JMS-SX102 spectrometer at a flow rate of 6 l/min by use of a microprocessor-controlled Pump 22 infusion syringe pump (Harvard Apparatus, Inc., South Natick, MA). The mobile phase composition was 0.1% trifluoroacetic acid, 30% methanol, and 0.3% 1-thioglycerol in water (v/v). The accelerating voltage was 10 kV, and ions were analyzed in the positive mode as a function of their m/z ratio.
Phosphorylation of Ca 2ϩ -ATPase-Phosphorylation of the SR vesicles (0.4 mg/ml) with 1 mM 32 P i was performed at 25°C for 10 min in 20 mM MgCl 2 , 5 mM EGTA, 40% (v/v) Me 2 SO, 30 mM MOPS, and 7 mM sodium borate (pH 7.0). The reaction was quenched with trichloroacetic acid containing P i . Phosphorylation of the vesicles (1 mg/ml) with 3 mM acetyl [ 32 P]phosphate was performed at 25°C in 5 mM MgCl 2 , 2 mM CaCl 2 , 50 mM KCl, 50 mM Tris, and 20 mM sodium borate (pH 8.0). The reaction was quenched with trichloroacetic acid containing nonradioactive acetyl phosphate. The amount of EP formed was determined as described previously (19).
Miscellaneous Methods-CHD was purchased from Aldrich. Girard's reagent T and KF were from Nacalai Tesque (Kyoto, Japan). ATP was from Yamasa Biochemicals (Choshi, Japan). Acetyl phosphate was from Kohjin (Tokyo, Japan). TPCK-treated trypsin and Na 3 VO 4 were from Sigma. [␥-32 P]ATP was obtained from NEN Life Science Products. Vanadate solutions were prepared from Na 3 VO 4 according to Goodno (25) just before use. 32 P i was purified according to Kanazawa and Boyer (12). Acetyl [ 32 P]phosphate was prepared by Procedure B in the method of Stadtman (28). Protein concentrations were determined by the method of Lowry et al. (29) with bovine serum albumin as a standard. Data were analyzed by the nonlinear least squares method according to the algorithm of Marquardt (30) as described previously (19). Fig. 1A, when the SR vesicles were pretreated with neither F Ϫ nor Mg 2ϩ and then treated with CHD, EP formation from P i was inhibited with pseudo-first order kinetics and fell to 13% of the original level in 90 min (É). When the vesicles were pretreated with F Ϫ and Mg 2ϩ and then treated with CHD, the enzyme was protected appreciably against the CHD-induced inhibition of EP formation (tightly bound F Ϫ and Mg 2ϩ were removed before the determination of EP formation) (E). The pretreatment without F Ϫ (Ⅺ) or Mg 2ϩ (Ç) provided no protection. The observed first order kinetics of the CHD-induced inhibition of EP formation (Fig. 1, A-C, solid lines) is consistent with the possibility that this inhibition was caused by modification of a single residue. However, it is also possible that more than one vital arginyl residue was modified at an approximately equal rate.

CHD-induced Inhibition of EP Formation from P i and Protection by Pretreatment with F Ϫ and Mg 2ϩ against the Inhibition-As shown in
Protection by Presence of Vanadate and Mg 2ϩ against CHDinduced Inhibition of EP Formation from P i -As shown in Fig.   1B, when the SR vesicles were treated with CHD in the presence of both vanadate and Mg 2ϩ , the enzyme was protected against the CHD-induced inhibition of EP formation from P i (bound vanadate and Mg 2ϩ were removed before the determination of EP formation) (E). In the absence of vanadate and presence of 20 mM Mg 2ϩ (Ⅺ) or in the presence of 0.5 mM vanadate and absence of Mg 2ϩ (Ç), no protection was observed.
Protection by EP Formation from P i against CHD-induced Inhibition of EP Formation from P i -As shown in Fig. 1C, when the SR vesicles were preincubated with P i in the presence of Mg 2ϩ and absence of Ca 2ϩ (4.1 nmol of EP/mg of SR vesicles was formed by this preincubation) and then treated with CHD, the enzyme was protected against the CHD-induced inhibition of EP formation from P i (E). However, this protection was somewhat less effective than the protection by the pretreatment with F Ϫ and Mg 2ϩ or by the presence of vanadate and Mg 2ϩ (compare Fig. 1C with Fig. 1, A and B). Binding of P i to the catalytic site in the presence of 150 mM P i and absence of Mg 2ϩ provided no protection (Ç).
CHD-induced Inhibition of Tight Binding of F Ϫ and Mg 2ϩ -The SR vesicles were treated with or without CHD for various times, and then tight binding of F Ϫ and Mg 2ϩ was determined (Fig. 2). The amounts of tightly bound F Ϫ and Mg 2ϩ at zero time of the treatment were 14.7 and 8.4 nmol/mg, respectively, which are 3.7 and 2.1 times the content of phosphorylation site, respectively (4.0 nmol/mg, see "Experimental Procedures"). This is in agreement with our previous conclusion (20) that the stoichiometry of tight binding of F Ϫ and Mg 2ϩ to the maximum level of phosphorylation is 4:2:1. The tight binding was inhibited progressively during the treatment with CHD, although the inhibition was considerably slower than the CHD-induced inhibition of EP formation from P i . This less effective inhibition may be in part due to the fact (18,19) that the binding of F Ϫ and Mg 2ϩ to the enzyme is virtually irreversible in the absence of Ca 2ϩ .
CHD-induced Inhibition of Vanadate Binding and Protection by Pretreatment with F Ϫ and Mg 2ϩ against the Inhibition-The SR vesicles were pretreated with or without F Ϫ in the presence of Mg 2ϩ and then treated with or without CHD for 90 min. After tightly bound F Ϫ and Mg 2ϩ were removed, vanadate binding was determined at various concentrations of vanadate in the presence and absence of Mg 2ϩ (Fig. 3A). In the control (Ç, å), in which the vesicles were pretreated without F Ϫ and then treated without CHD, Mg 2ϩ -dependent vanadate binding (broken line) increased with increasing concentrations of vanadate and was saturated with 10 M vanadate. The maximum level of this binding was in substantial agreement with the content of the phosphorylation site. The Mg 2ϩ -dependent binding was almost completely inhibited by the treatment with CHD (Ⅺ, f). When the vesicles were pretreated with F Ϫ and Mg 2ϩ and then treated with CHD, the enzyme was appreciably protected by this pretreatment against the CHD-induced inhibition of Mg 2ϩdependent vanadate binding (E, q). When the vesicles were pretreated without F Ϫ and then treated with CHD for various times, vanadate binding was inhibited with pseudo-first order kinetics at a rate of 0.028 min Ϫ1 (Fig. 3B, line drawn through  open squares). This rate of inhibition was in good agreement with the rate (0.023-0.024 min Ϫ1 ) of the CHD-induced inhibition of EP formation from P i under the same conditions (see solid lines in Fig. 1, A and B). The inhibition of vanadate binding was slowed by the pretreatment with F Ϫ and Mg 2ϩ (Fig. 3B, line drawn through open circles) to a similar extent as the inhibition of EP formation from P i (cf. Fig. 1A).
Lack of CHD-induced Inhibition of EP Formation from Acetyl Phosphate-The SR vesicles were pretreated with or without F Ϫ in the presence of Mg 2ϩ . In the experiments shown in Fig.   FIG. 3. CHD- 4A, the pretreated vesicles were treated with or without CHD for various times. After tightly bound F Ϫ and Mg 2ϩ were removed, EP formation from acetyl phosphate was determined. CHD caused no substantial inhibition of the EP formation when the vesicles were pretreated without F Ϫ (compare Ⅺ with f). The pretreatment with F Ϫ and Mg 2ϩ exerted a slight protective effect during the control incubation without CHD (compare q with f). In the experiments shown in Fig. 4B, the pretreated vesicles were treated with or without CHD for 90 min and then phosphorylated with acetyl phosphate for various times. When the vesicles were pretreated without F Ϫ and treated with CHD, the rate of EP formation was essentially the same as that of EP formation in native SR vesicles (compare e with ϫ). The pretreatment with F Ϫ and Mg 2ϩ had virtually no effect on the kinetics of EP formation in the CHD-treated vesicles (compare E with Ⅺ). When the vesicles were pretreated without F Ϫ and treated without CHD, EP formation was appreciably faster than that in the native vesicles for unknown reasons (compare f with ϫ).
Peptide Mapping of Tryptic Digests of CHD-modified SR Vesicles-In the first series of experiments (Fig. 5, A-D), the SR vesicles were pretreated with neither F Ϫ nor Mg 2ϩ (A), with F Ϫ and without Mg 2ϩ (B), without F Ϫ and with Mg 2ϩ (C), or with F Ϫ and Mg 2ϩ (D) and then treated with CHD. After tightly bound F Ϫ and Mg 2ϩ were removed, the vesicles were digested with TPCK-treated trypsin and subjected to reversed phase HPLC. The peptide maps at 214 nm of the digests (Fig. 5, A-D In the third series of experiments (Fig. 5, H-J), the SR vesicles were preincubated in the absence of P i and presence of 8.33 mM MgCl 2 (H), in the presence of 156 mM P i and absence of Mg 2ϩ (I), or in the presence of 1.04 mM P i and 8.33 mM MgCl 2 (J) and then treated with CHD. The vesicles were digested with TPCK-treated trypsin and subjected to reversed phase HPLC. The preincubation in the presence of P i and Mg 2ϩ (4.1 nmol of EP per mg of SR vesicles was formed by this preincubation) induced a small but definite reduction in the peaks indicated by arrows (compare upper trace of Fig. 5J with upper trace of Fig.  5H). Binding of P i to the catalytic site in the presence of 150 mM P i and absence of Mg 2ϩ had no effect on these peaks (compare upper trace of Fig. 5I with upper trace of Fig. 5H).
It is clear from the chromatographic profiles that the peaks reduced by the presence of vanadate and Mg 2ϩ (Fig. 5, E-G,  arrows) and by the presence of P i and Mg 2ϩ (Fig. 5, H-J, arrows) corresponded to those reduced by the pretreatment with F Ϫ and Mg 2ϩ (Fig. 5, A-D, arrows).
Purification of Peptides Containing DHCH-Arg-The SR vesicles were pretreated without F Ϫ and with Mg 2ϩ (Fig. 6A) or with F Ϫ and Mg 2ϩ (Fig. 6B) and then treated with CHD. After tightly bound F Ϫ and Mg 2ϩ were removed, the vesicles were digested with TPCK-treated trypsin and subjected to the first reversed phase HPLC (Fig. 6, A and B).
Fraction I and fraction II in the figures were pooled separately and subjected to the second reversed phase HPLC (Fig.  6, C--F). Fraction I in Fig. 6A gave a single major peak at 325 nm (upper trace of Fig. 6C, fraction I). Fraction I in Fig. 6B gave a major peak at 325 nm at the same retention time as in Fig. 6C (upper trace of Fig. 6D, fraction I), but this peak was much smaller than the corresponding peak in Fig. 6C. Fraction II in Fig. 6A gave two close peaks at 325 nm (upper trace of Fig. 6E, fraction II). These two peaks were greatly reduced (upper trace of Fig. 6F, fraction II) when HPLC was performed with fraction II in Fig. 6B.
Peptides containing DHCH-Arg in fraction I in Fig. 6 (C and D) and fraction II in Fig. 6 (E and F) were further purified by the third reversed phase HPLC (Fig. 6, G-J). Fraction I in Fig.  6C gave a major peak and a minor peak at 325 nm (upper trace of Fig. 6G, peak Ia and peak Ib). Fraction I in Fig. 6D also gave a major peak and a minor peak at 325 nm (upper trace of Fig.  6H, peak Ia and peak Ib), but these peaks were much smaller than the corresponding peaks in Fig. 6G. Fraction II in Fig. 6E gave two major peaks at 325 nm (upper trace of Fig. 6I, peak IIa and peak IIb). Fraction II in Fig. 6F also gave two major peaks at 325 nm (upper trace of Fig. 6J, peak IIa and peak IIb), but these two peaks were much smaller than the corresponding peaks in Fig. 6I.
Sequencing and Mass Analysis of Peptides Containing DHCH-Arg-The peptides containing DHCH-Arg isolated as above (peak Ia and peak Ib in Fig. 6, G and H, and peak IIa and peak IIb in Fig. 6, I and J) were sequenced (Table I). All the isolated peptides gave the same sequence (His-190 to Lys-204) in the Ca 2ϩ -ATPase, in which a missing residue (X) corresponded to Arg-198. These results suggest that Arg-198 in the Ca 2ϩ -ATPase was modified with CHD. This is consistent with the previously reported findings (16)  to tryptic cleavage. The molecular masses of these peptides determined by mass spectrometry were in good agreement with the monoisotopic mass calculated from the above sequence on the assumption that X is DHCH-Arg (Table II). These findings show that Arg-198 in the Ca 2ϩ -ATPase was modified with CHD and that this modification of Arg-198 was strongly inhibited by the pretreatment with F Ϫ and Mg 2ϩ . Furthermore, it is very likely that the modification of Arg-198 with CHD was specifically inhibited also by the presence of vanadate and Mg 2ϩ (Fig.   5, E-G, arrows) or by the presence of P i and Mg 2ϩ (Fig. 5, H-J,  arrows). The reason why peptides with the same sequence and same modification gave different retention times in HPLC remains obscure. DISCUSSION The observed protection by tight binding of F Ϫ and Mg 2ϩ (Fig. 1A) or by high affinity binding of vanadate (Fig. 1B) against the CHD-induced inhibition of EP formation from P i FIG. 6. Purification of DHCH-Arg-containing peptides that are sensitive to pretreatment with F ؊ and Mg 2؉ . The SR vesicles were pretreated without KF and with MgCl 2 (A), or with KF and MgCl 2 (B), otherwise as described under "Experimental Procedures." The vesicles were then incubated with CHD, digested with TPCK-treated trypsin, and subjected to the first reversed phase HPLC (A and B), as in Fig. 5 (A-D). Fractions of 0.3 ml each were collected, and fractions indicated by horizontal bars (I and II) were pooled separately. Fractions I and II from A and B were subjected to the second reversed phase HPLC (C-F). Fractions subjected to HPLC in C, D, E, and F were fraction I from A, fraction I from B, fraction II from A, and fraction II from B, respectively. Fractions of 0.7 ml each were collected, and fractions indicated by horizontal bars (I and II) were pooled separately. Elution was performed with the following linear gradients of acetonitrile in 1 mM acetic acid: C and D, 0% from 0 to 10 min, 9% at 20 min, 17.5% at 105 min, and 100% at 115 min; E and F, 0% from 0 to 10 min, 10% at 20 min, 18.5% at 105 min, and 100% at 115 min. Fractions I and II from C-F were subjected to the third reversed phase HPLC (G-J). Fractions subjected to HPLC in G, H, I, and J were fraction I from C, fraction I from D, fraction II from E, and fraction II from F, respectively. Fractions of 0.7 ml each were collected in G and H, and fractions of 1.4 ml each were collected in I and J. Elution was performed with the following linear gradients in 0.1% trifluoroacetic acid: G and H, 0% at 0 min, 12% at 10 min, 15.5% at 80 min, and 100% at 90 min; I and J, 0% at 0 min, 10% at 10 min, 13.9% at 140 min, and 100% at 150 min. Fractions indicated by horizontal bars (Ia, Ib, IIa, and IIb) were pooled separately for sequencing and mass spectrometry. In A-J, the absorbance at 214 nm (lower traces) and the absorbance of peptides containing DHCH-Arg at 325 nm (upper traces) were monitored.
suggests that an arginyl residue(s) protected by these transition state analogs contributes toward formation of the transition state for EP formation from the magnesium-enzyme-phosphate complex or, alternatively, that the protected arginyl residue(s) is located close to the essential components (bound phosphate, bound Mg 2ϩ , or other functional groups) of this transition state. This view is consistent with the findings that binding of the transition state analogs is inhibited by the modification with CHD ( Figs. 2 and 3) and that the enzyme is protected by tight binding of F Ϫ and Mg 2ϩ against the CHDinduced inhibition of high affinity vanadate binding (Fig. 3). It is also in harmony with the observed lack of protection by the enzyme-magnesium (Fig. 1B) and enzyme-phosphate (Fig. 1C) complexes, because the structures of these complexes should be different from the structure of the transition state for EP formation from the magnesium-enzyme-phosphate complex.
The reduction in two specific peaks by tight binding of F Ϫ and Mg 2ϩ (Fig. 5D, arrows) or by high affinity binding of vanadate (Fig. 5G, arrows) indicates that a few specific arginyl residues are selectively protected by the transition state analogs against the modification with CHD. The data from peptide purification (Fig. 6), sequencing (Table I), and mass spectrometry (Table II) show that the only major arginyl residue protected by these analogs is Arg-198. These findings indicate that modification of Arg-198 with CHD is responsible for at least a part (the portion reduced by the transition state analogs) of the CHD-induced inhibition of EP formation from P i (Fig. 1, A and  B). Dux and Martonosi (31) showed previously that tryptic cleavage of the Ca 2ϩ -ATPase at Arg-198 is inhibited by high affinity binding of vanadate to the enzyme. Their findings appear to be compatible with our present observations. The lack of protection by the presence of 20 mM Mg 2ϩ alone against the CHD-induced inhibition of EP formation from P i (Fig. 1B) and against the modification of Arg-198 with CHD (Fig. 5E, arrows) indicates that Arg-198 is not located at the Mg 2ϩ binding site in the enzyme-magnesium complex. In addition, the lack of protection by the presence of 150 mM P i alone against the CHD-induced inhibition of EP formation from P i (Fig. 1C) and against the modification of Arg-198 with CHD (Fig. 5I, arrows) indicates that Arg-198 is not located at the P i binding site in the enzyme-phosphate complex. The slightly less effective protection by EP formation from P i against the CHD-induced inhibition of EP formation from P i (Fig. 1C) and against the modification of Arg-198 with CHD (Fig. 5J, arrows) suggests that Arg-198 is located close to, but outside, the phosphorylation site. The location of Arg-198 in proximity to the phosphorylation site is consistent with the previously reported findings (32) that tryptic cleavage of the Ca 2ϩ -ATPase at Arg-198 is inhibited by EP formation from P i .
No appreciable effects of the modification with CHD on the extent (Fig. 4A) and rate (Fig. 4B) of EP formation from acetyl phosphate indicate that phosphoryl transfer from the bound substrate to the phosphorylation site in the forward reaction is not substantially affected by the modification of Arg-198. This finding also shows definitely that Arg-198 is located outside the phosphorylation site.
Chen et al. (33) showed previously that binding of ␥,␤-bidentate CrATP to the catalytic site of the Ca 2ϩ -ATPase is inhibited by P i or vanadate in the presence of Mg 2ϩ , that the chromium moiety of the bound CrATP accepts a ligand from the enzyme to form an exchange inert coordination complex, and that the site of CrATP attachment through this stable coordinate bond is situated on the A 2 (32) tryptic fragment (Ac-Met-1 to Arg-198). Their observations are consistent with our present view that Arg-198 is located in or close to the catalytic site in the transition state for EP formation from magnesium-enzyme-phosphate complex.
Arg-198 is conserved in sarco(endo)plasmic reticulum Ca 2ϩ -  Fig. 6 (G-J) were sequenced. The numbers in parentheses indicate pmol of amino acid at the given cycle. X represents the expected DHCH-arginyl residues, which cannot be detected by the method of sequencing used. The dash (-) indicates that no phenylthiohydantoin derivatives were detected.

Mass of peptides containing DHCH-Arg
Masses of the peptides containing DHCH-Arg in Table I were determined. Calculations of monoisotopic mass are based on the assumption that X is DHCH-Arg. MH ϩ , monoisotopic mass. ATPases (1, 34 -36) but not in other P-type ATPases including the Na ϩ ,K ϩ -ATPase (37,38), H ϩ ,K ϩ -ATPase (39,40), and plasma membrane Ca 2ϩ -ATPase (41). Therefore, although it is possible that Arg-198 is directly involved in the catalytic process of EP formation from P i , it appears more likely that the observed inhibition (at least the portion reduced by the transition state analogs) is due to steric hindrance induced by CHDmodified Arg-198. The secondary structural model for the Ca 2ϩ -ATPase suggests that the enzyme is composed of 10 transmembrane ␣-helices (M 1 -M 10 ) and a cytoplasmic globular fraction that is divided into two main domains, a small cytoplasmic loop (Ala-132-Asp-237 between M 2 and M 3 ) and a large cytoplasmic loop (Asn-330 -Phe-740 between M 4 and M 5 ) (1). The large cytoplasmic loop contains the phosphorylation site and the ATP binding site. The functional role of the small cytoplasmic loop is less clear. However, in the previous studies on the function of this region, it was shown that the conformational change of EP is blocked by site-directed mutations of Thr-181, Gly-182, Glu-183 (42), and Gly-233 (43) and by a tryptic cleavage at Lys-234 or Arg-236 (44). It was also shown (43) that the affinity for P i in EP formation from P i is reduced in the Gly-233 mutants. Accordingly, it is likely that this small cytoplasmic loop including Arg-198 also contributes in part to the catalytic site.