The Inactivation of the Acyl Phosphatase Activity Catalyzed by the Sulfenic Acid Form of Glyceraldehyde 3-Phosphate Dhd e y rogenase by Dimedone and Olefins*

SUMMARY Treatment of the sulfenic acid form of glyceraldehyde 3-phosphate dehydrogenase (EC 1.2.1.12) with a Z-fold molar excess of dimedone over the concentration of enzyme subunit completely inactivates the acyl phosphatase reaction catalyzed by the oxidized enzyme. The dehydrogenase activity catalyzed by the reduced form of the enzyme is not reacti-vated when the dimedone-inactivated enzyme is treated with dithiothreitol. When the acyl phosphatase is inactivated by (“Cldimedone -1 c(g atom of 14C is incorporated per peq of oxidized enzyme subunit which is not removed by gel filtration on Sephadex G-25. A 14C-labeled peptide from a tryptic digest of the acyl phosphatase which was inactivated with [14C]dimedone has been isolated in pure form and has been subjected to sequence analysis. This analysis has shown that [“CJdimedone forms a thioether derivative of Cys-149 by reacting with the sulfenic acid at the active site of the acyl phosphatase. The dehydrogenase activity is unaffected by [“Cldimedone when reduced glyceraldehyde 3-phosphate dehydrogenase is treated with [‘4C]dimedone and 14C radioactivity

SUMMARY Treatment of the sulfenic acid form of glyceraldehyde 3phosphate dehydrogenase (EC 1.2.1.12) with a Z-fold molar excess of dimedone over the concentration of enzyme subunit completely inactivates the acyl phosphatase reaction catalyzed by the oxidized enzyme. The dehydrogenase activity catalyzed by the reduced form of the enzyme is not reactivated when the dimedone-inactivated enzyme is treated with dithiothreitol. When the acyl phosphatase is inactivated by ("Cldimedone -1 c(g atom of 14C is incorporated per peq of oxidized enzyme subunit which is not removed by gel filtration on Sephadex G-25. A 14C-labeled peptide from a tryptic digest of the acyl phosphatase which was inactivated with [14C]dimedone has been isolated in pure form and has been subjected to sequence analysis. This analysis has shown that ["CJdimedone forms a thioether derivative of Cys-149 by reacting with the sulfenic acid at the active site of the acyl phosphatase. The dehydrogenase activity is unaffected by ["Cldimedone when reduced glyceraldehyde 3-phosphate dehydrogenase is treated with ['4C]dimedone and 14C radioactivity is not incorporated into the reduced protein. The acyl phosphatase activity catalyzed by the sulfenic acid form of glyceraldehyde j-phosphate dehydrogenase isolated from pig muscle is also inactivated by 25 mrd 3-cyclohexene-lcarboxylate at pH 5.1 in the presence of 0.1 M (NH4)S04 and other salts. Treatment of the inactivated acyl phosphatase with dithiothreitol does not reactivate the dehydrogenase reaction catalyzed by the reduced form of the enzyme. This indicates that treatment of the sulfenic acid form of the enzyme with the olefin leads to the covalent modification of Cys-149, the catalytically essentially sulfhydryl group for the dehydrogenase. Treatment of reduced glyceraldehyde 3-phosphate dehydrogenase with 3-cyclohexene-l-carboxylate under identical conditions has no effect on the dehydrogenase activity. The acyl phosphatase is also inactivated by 5.0 mM dihydropyran at pH 7.6 and by 30 mr.r tetrahydrophthalimide at pH 6.0. Treatment of the acyl phosphatase * This work was supported by research grants from the United States Public Health Service, National Institute of General Medical Sciences 974) and from the National Science Foundation .
$ Recipient of a study grant from the National Science Development Board of the Philippines. 5 To whom correspondence should be addressed.
inactivated by these reagents with dithiothreitol does not reactivate the dehydrogenase activity catalyzed by the reduced form of the enzyme. Under identical conditions used to inactivate the acyl phosphatase, these two olefins do not inactivate the dehydrogenase activity catalyzed by native glyceraldehyde 3-phosphate dehydrogenase when added to it directly. Amino acid sequence analysis of the tryptic peptide which contains Cys-149 after the acyl phosphatase was inactivated with tetrahydrophthalimide suggests that an adduct between the sulfenic acid at the active site of the acyl phosphatase and tetrahydrophthalimide is formed during the inactivation.
The oxidation of the sulfhydryl group at the active site of glyceraldehyde 3-phosphate dehydrogenase to a sulfenic acid converts the dehydrogenase to an acyl phosphatasc (l-3).
Evidence has been presented which suggests that the sulfenic acid residue participates directly in t,he hydrolytic activity as described by Equations 1 and 2.
i; 0 A number of nucleophiles which are classical carbonyl reagents inactivate the acyl phosphatase activity of the enzyme by forming sulfenyl derivatives of cystcine residue 149. Among these are cyanide, bisulfite, semicarbaeide, and hydrazines (2,4,5). Sulfenyl halides arc known to react with active methylene compounds to form the corresponding thiocthers (6). It was therefore of interest to see if a protein sulfenic acid would also react with active methylene compounds.
Expcrimcntal evi dence is presented which supports such a reaction of [r4C]dimedone with the sulfenic acid form of glyceraldehyde 3-phosphate dehydrogenase.
Sulfenyl halides and ot,her sulfenyl derivatives such as sulfenyl carboxylates are known to add across olefinic bonds by a reaction similar to the addition of Hrz across olefinic bonds. The addition of sulfenyl compounds to olefins proceeds with the formation of an episulfonium intermediate as shown by Equation 3.
The product of the reaction is invariably the trans-P-substituted thioether where X-or a component of the solvent reacts with the episulfonium intermediate (7,8). We wish to report that water-soluble olefins also irreversibly inactivate the acyl phosphatase reaction catalyzed by the sulfenic acid form of glyceraldehyde a-phosphate dehydrogenase. In addition evidence is presented that tetrahydrophthalimide reacts with the sulfenic acid form of the enzyme to form a covalent derivative of Cys-149.

Materials
Glyceraldehyde 3-phosphate dehydrogenase was prepared from pig muscle as described by Elodi and Szijrenyi (9) with a modification described previously (10). The enzyme was oxidized to the sulfenic acid form in the following manner.
Crystals of the dehydrogenase were collected by centrifugation and were dissolved in 20 mM Verona], pH 7.6, to give a final protein concentration of 20 to 25 mg per ml. A neutralized solution of o-iodosobenzoate was added to the dissolved enzyme in the presence of 10 mM NAD+ to a final concentration of 20 mM. Oxidation was carried out for 1 hour at 0' at which time excess reagents were removed by gel filtration on a column (2.0 X 30 cm) of Sephadex G-25 (medium) which was equilibrated and eluted with 20 mM Verona], pH 7.6. The concentrated protein fractions were combined and maintained at 0' until they were used in the various experiments.
All experiments were performed within 6 hours after the isolation of the oxidized enzyme.
[4J*C]Dimedone was generously provided by Dr. David Sigman of the University of California, Los Angeles, School of Medicine. Spectrophotometric grade nitromethane was purchased from the Aldrich Chemical Co. and was used without further purification. Twice recrystallized trypsin was purchased from Worthington and was treated with L-1-tosylamido-2-phenylethyl chloromethyl ketone as described by Carpenter (11). Bacterial protease type VII (subtilisin) was purchased from Sigma Chemical Co.
3-Cyclohexene-1-carboxylic acid was purchased from Aldrich Chemical Co. and was redistilled twice under reduced pressure before use. Dihydropyran was purchased from Aldrich Chemical Co. and was redistilled immediately before use. cis-A-4-Tetrahydrophthalimide was purchased from Aldrich Chemical Co. and was recrystallized from benzene before use. ]I-"C]Iodoacetic acid was purchased from New England Nuclear and was dissolved in 0.1 M NaOH to a final concentration of 0.1 M. Before use it was extracted with CC14 to remove traces of I1 which were present as a contaminant.
Assays-The acyl phosphatase activity was assayed by the method of Ehring and Colowick (1). The dehydrogenase activity was assayed as described previously (12). Radioactivity was 6235 monitored with the use of a Beckman LS 100 scintillation counter with the scintillation fluid described by Bray (13).

Enzyme with [W]Dimedone
Oxidized pig muscle glyceraldehyde 3-phosphate dehydrogenase (140 mg or 4 pmoles of subunit) was treated with 8 pmoles of [i4C]dimedone in 9.61 ml of 20 mM Verona1 buffer, pH 7.6. The specific radioactivity of the ['"Cldimedone was 1.25 X lo6 cpm per rmole. The acyl phosphatase activity in the reaction mixture disaopeared after 50 min at room temper&me at which time it was gel:filtered on a column (2.0 X 30 cm) of Senhadex G-25 (medium) which was equilibrated and eluted with 20~rn~ Verona1 buffer, pH 7.6. The radioactive protein fractions were combined and carboxymethylated with 200 rmoles of neutralized iodoacetate in the presence of 8 M urea. Thiodiglycol was added to a final concentration of 0.1% and the reaction mixture was dialyzed exhaustively against 1 mM HCI-0.1% thiodiglycol.

Peptide
The dialyzed radioactive protein was digested with 2.5 mg of trypsin in the presence of 0.5'% NHdHCOa for 6 hours at 37". The tryptic digest was then lyophilized.
Step I: Fractionation of Sephadex G-%-The freeze-dried tryptic digest was dissolved in 10 ml of 0.05 M NHdOH-0.1% thiodiglycol and gel-filtered on a column (3.0 X 100 cm) of Sephadex G-25 (fine) which was equilibrated and eluted with 0.05 M NHaOH-0.1% thiodiglycol and collected in 5.5-m] fractions.
Fractions 43 to 53, which contain all of the radioactivity as shown in Fig. 1 were combined and lyophilized.
Step ZZ: Fractionation of Sephadex G-60-The lyophilized radioactive material from the previous step was dissolved in 5 ml of 0.05 M NHdOH-0.1% thiodiglycol and gel-filtered on a Sephadex G-50 column (3.0 X 95 cm). Fractions 45 to 68, which contained the greatest amount of radioactivity, were combined and lyophilized. This fractionation is shown in Fig. 2.
Step ZZZ: Column Chromatography on DEAE-Sephadex A-,%- The radioactive material from Step II was dissolved in 3.0 ml of the starting buffer and loaded onto a DEAE-Sephadex A-25 column (1.6 X 20 cm) which was equilibrated with 1% pyridine, 1% collidine, 0.3% thiodiglycol, which was adjusted to pH 8.0 with acetic acid. The column was eluted with a pH gradient.
The mixing chamber contained 600 ml of the starting buffer, pH 8.0, and the reservoir chamber contained 600 ml of 1% pyridine, 1% collidine, 0.3% thiodiglycol which was adjusted to pH 6.5 with acetic acid. The effluent was collected in 5.5-m] fractions.
The elution pattern is shown in Fig. 3. The radioactive Fractions 79 to 90 were combined and lyophilized.
Step IV: Desalting by Gel Filtration on Sephadex G-%-The partially dried fractions from Step III were dissolved in 3.0 ml of 0.05 M NHIOH-0.17s thiodiglycol and desalted by gel filtration on column (3.0 X 100 cm) and lyophilized. Fractionation on Sephadex G-50 of the radio-as described under "Experimental Procedure" ( l ). Fifty-microactive fraction obtained from Sephadex G-25 gel filtration.
The lyophilized radioactive Fractions 43 to 53 illustrated in Fig. 1 were liter samples of the collected fractions were assayed at 750 nm by the method of Lowry et al. (14).
The purity of the [%]dimedone-labeled active site tryptic peptide was monitored by high voltage electrophoresis on Whatman No. 3MM paper using 1% (NH&COa, pH 8.9, as electrolyte.

Amino Acid Analysis
Peptide samples for amino acid analysis were prepared as follows. About 15 to 25 nmoles of the peptide in 1.0 ml of 0.05 M NH40H-0.1a/o thiodiglycol were extracted with 1.0 ml of benzene three times, to remove the thiodiglycol.
The aqueous peptide solution was then lyophilized and then hydrolyzed with 300 ~1 of constant boiling HCl in vacuum-sealed tubes at 110" for 24 or 48 hours. When thiodiglycol was not present about 15 to 25 nmoles of peptide were lyophilized and were hydrolyzed with 300 $1 of constant boiling HCl in vacuum-sealed tubes at 110' for 24 or 48 hours. Analyses were conducted with the use of a Beckman model 119 automatic amino acid analyzer (15).

RESULTS
Inactivation of Acyl Phosphatase Activity by Dimedone and Nitromethune- Table I shows that the acyl phosphatase reaction catalyzed by oxidized glyceraldehyde 3-phosphate dehydrogenase is inactivated by fairly low concentrations of dimedone. For instance, when a stoichiometric concentration of dimedone is added to the oxidized enzyme at room temperature, 55% of the acyl phosphatase is lost in 30 min. Fig. 4 shows the time dependence of the inactivation of the acyl phosphatase activity when the oxidased enzyme is treated with a a-fold molar excess of dimedone over the concentration of enzyme subunit. The oxidized enzyme is completely inactivated by dimedone within 1 hour at room temperature while the oxidized enzyme in an untreated control lost approximately 15% of its activity during a similar incubation. Treatment of the control reaction mixture at the end of l-hour incubation with 0.01 M dithiothreitol for 30 min at room temperature led to the reactivation of 90% of the dehydrogenase activity catalyzed by reduced glyceraldehyde 3-phosphate dehydrogenase. No dehydrogenase activity was recovered when the reaction mixture which contained dimedone was treated with dithiothreitol. This suggests that treatment of the sulfenic acid form of glyceraldehyde 3-phosphate dehydrogenase with dimedone leads to the covalent modification of Cys-149. When reduced glyceraldehyde 3-phosphate dehydrogenase is treated with 1 mM dimedone, the dehydrogenase activity is not affected. This also suggests that dimedone reacts with the sulfenic acid at the active site of the acyl phosphatase.
Nitromethane, another active methylene compound with a much higher pK, than dimedone, inactivates the acyl phosphatase activity catalyzed by the sulfenic acid form of glyceraldehyde 3-phosphate dehydrogenase but much less effectively thad dimedone. For instance a a-fold molar excess of dimedone over Inactivation of acyl phosphatase activity of oxidized glyceraldehyde S-phosphate dehydrogenase by dimedone The reaction mixtures contained in 1.0 ml of 20 mM Verona1 buffer, pH 7.6, oxidized enzyme, 0.10 peq, and dimedone in the concentrations indicated.
After 30 min at room temperature, O.lO-ml samples of the reaction mixtures were assayed for acyl phosphatase activity as described by Ehring and Colowick (1). The reaction mixture contained in 1.0 ml of 20 mM Veronal, pH 7.6, 0.1 req of oxidized enzyme subunit, and 0.2 Fmole of dimedone was incubated at room temperature.
At the times indicated O.l-ml samples were withdrawn and assayed for acyl phosphatase activity as described by Ehring and Colowick (1). l , the reaction mixture which contained dimedone; 0, an untreated control reaction mixture.
the concentration of enzyme subunit completely inactivates the acyl phosphatase after an hour, while a 200-fold molar excess of nitromethane results in only 30y0 inactivation of the acyl phosphatase under the same conditions.

Covalent Modification of Oxidized Glyceraldehyde S-Phosphate
Dehydrogenase by [14C]Dimedone- Table II shows that ['"Cldimedone is incorporated into the sulfenic acid form of glyceraldehyde 3-phosphate dehydrogenase during the inactivation of the acyl phosphatase reaction. With the exception of the results of Experiment 3 in Table II, nearly 1 g atom of [14C]dimedone was incorporated per mole of enzyme subunit during the inactivation of the acyl phosphatase activity. Experiment 5 in Table II Table II was digested with trypsin in the presence of thiodiglycol.
The tryptic digest was fractionated as described under "Experimental Procedure" by means of sequential column chromatography on Sephadex G-25 (Fig. l), Sephadex G-50 (Fig. 2), and DEAE-Sephadex A-25 (Fig. 3). All of the columns were developed in the presence of thiodiglycol which was introduced as an antioxidant (16). After the DEAE-Sephadex step, the radioactive material in Fractions 79 to 90 shown in Fig. 3 was desalted on a column (3.0 X 100 cm) of Sephadex G-25 which was equilibrated and eluted with 0.05 M NHdOH which contained 0.1% thiodiglycol. After desalting, a small fraction of the single radioactive peak which appeared in the void volume was sub- to paper electrophoresis at pH 8.9 for 2 hours at 1.5 kv. A radioautograph of this ionogram revealed a major radioactive band and a minor radioactive band which migrated further than the major band. When the ionogram was stained with ninhydrin-cadmium reagent (17) only the material which migrated with the major radioactive band was ninhydrin-positive.
The amino acid composition of the radioactive peptide which wan obtained after the desalting step is shown in Table III.
Comparison of the amino acid composition of the 14C-labeled tryptic peptide with the known amino acid composition of the tryptic peptide of pig glyceraldehyde S-phosphate dehydrogenase which contains Cys-149 (18) strongly indicates that [14C]dimedone reacts covalently with Cys-149 when it inactivates the acyl phosphatase.
Acid hydrolysates of the tryptic peptide labeled with ["Cldimedone contained significant amounts of cysteic acid as shown in Table III.
No other unusual amino acids were detected on the chromatograms.
It is possible that the dimedone derivative of Cys-149 is unstable to acid hydrolysis and that one of the degradation products is cysteic acid.
Sequence Analysis Which Proves That [14C]Dimedone Forms Derivative of Cys-143 When It Inactivates Acyl Phosphatase-The 14C-labeled tryptic peptide which was isolated in the presence of thiodiglycol was digested with 50 pg of subtilisin for 4 hours in 0.5% NH4HC0, in the presence of 0.1% thiodiglycol. The subtilisin digest was freeze-dried and subjected to paper electrophoresis at pH 3.5 for 50 min at 3 kv. A radioautograph of the ionogram is shown in Fig. 5. The major radioactive subtilisin peptide, S4, which was basic at pH 3.5 was purified by subsequent paper electrophoresis at pH 6.5 and 8.9. The amino acid composition of the peptide Sd purified in this manner is shown in Table III. Based on specific radioactivity, 30 nmoles of peptide Sq were subjected to sequence analysis by the dansyli-Edman procedure of Gray (19) as shown in Table IV. The dansyl-amino acids were identified on polyamide thin layer sheets described by Woods and Wang (20) with the solvent systems described by Hartley (21). The NHz-terminal amino acid was identified as dansyl-alanine and after the first Edman step the new NH2terminal amino acid was identified as dansyl-serine.
To prove unequivocally that alanine was the NHz-terminal amino acid and serine was the new NH&erminal amino acid after the first Edman step, three separate dansyl steps were made in each case. This consumed a considerable portion of the radioactive peptide. Before the second Edman step, 8000 cpm of i4C remained.
No dansyl-amino acid, including dansyl-cysteic acid was identified after the second Edman step. The third Edman step removed 90% of the remaining radioactivity, indicating that the 3rd residue in the sequence of peptide S4 was labeled with ['"Cldimedone.
After the third Edman step, dansyl-threonine was identified as the fourth amino acid in the pentapeptide.
The sequence of the pentapeptide Se is underlined in the sequence of peptide from which it was derived as illustrated below. Ile-Val-Ser-Asn-Ala-Ser-~ys-Thr-Thr-Asn-Cys-Leu-Ala-Pro-Leu-Ala-Lys Cys-149 is marked with an asterisk and was shown to be labeled with [r4C]dimedone by this sequence analysis. Removal of [14C]Dimedone from Peptide S4 by Perjormic Acid Oxidation-A sample of peptide S4 which contained 3.1 x lo4 cpm of 14C was applied as a 1.5~cm band on a strip of Whatman No. 3MM paper which was exposed to performic acid vapor in a vacuum desiccator for 2 hours at room temperature.
The paper strip containing the performic acid-oxidized sample of peptide Sq was sewn onto a full sheet of Whatman No. 3MM paper. Untreated peptide S4 which contained 3.1 x lo4 cpm of 14C was applied to the paper along a common origin as a 1.5~cm band.
[i4C]Dimedone which contained 1.0 x lo4 cpm was applied along the common origin as an 0.5-cm band. The paper was then subjected to paper electrophoresis at pH 4.0 for 1 hour at 2 kv. A radioautograph of this ionogram is shown in Fig. 6. The untreated radioactive S4 was acidic at pH 4.0 while the radioactivity of the performic acid-treated S4 migrated with the neutral amino acids by endosmosis toward the cathode.
The migrated as two spots, one of which had the same mobility as the radioactive material released from S4 which was treated with performic acid. It is not known whether this material is dimedone or an oxidation product.
Although the same amount of radioactivity was applied per cm for the performic acid-treated S4, the untreated S4 and the [14C]dimedone, the untreated 14Clabeled S4 has a much more intense spot on the radioautogram. We attribute this to the extraction of ['4C]dimedone or the oxidized derivative into the Varsol which was used to dissipate the heat in our electrophoresis apparatus.
When the ionogram was stained with the ninhydrin-cadmium reagent, ninhydrin-positive material appeared under the radioactive spot on the radioautogram, under the untreated peptide Sd, and a ninhydrin-positive spot was detected where the performic acid-treated peptide S4 was applied which was more acidic than the untreated peptide S4. These experiments also suggest that [i4C]dimedone formed a covalent derivative of Cys-149.
Inactivation of Acyl Phosphatase Activity Catalyzed by Oxidized Glyceraldehyde S-Phosphate Dehydrogenase with S-Cyclohexene-l-Carboxylate, Dihydropyran, and Tetrahydrophthalimid- Table  V shows that incubation of oxidized glyceraldehyde 3-phosphate dehydrogenase (3.6 mg per ml) with 25 mM 3-cyclohexene-l-carboxylate at pH 5.1 in the presence of 0.1 M (NH&S04 leads to 94y0 inactivation of the acyl phosphatase within 15 min at room temperature when this activity is assayed at pH 7.6 (1). However, this olefin does not inactivate the acyl phosphatase at pH 5.1 in the absence of (NH&S04 and in fact it appears to stabilize the oxidized enzyme in the absence of the salt. The addition of 0.1 M (NH&S04 to the oxidized enzyme in the absence of 3-cyclohexene-I-carboxylate also stabilized the oxidized enzyme slightly at pH 5.1 when compared to the untreated control as shown in Table V. The reaction mixtures which were incubated under the various conditions shown in Table V were prepared in a manner to insure that each had a final pH of 5.1 as determined by a micro glass electrode. Therefore, differences in pH are not responsible for the results shown in Table V. At pH values greater than 5.1, the acyl phosphatase was progressively less sensitive to inactivation by 3-cyclohexene-l-carboxylate in the presence of 0.1 M (NH4)&04.
At pH 6.0 and above 25 mM 3-cyclohexene-l-carboxylate does not inactivate the acyl phosphatase in the presence or absence of added (NH&S04.
Table V also shows that the inactivation of the acyl phos- This suggests that the in-activa4ion of the acyl phosphatase by the olefin under these conditions prevents the reduction of Cys-149 to the sulfhydryl oxidation state which is required for the dehydrogenase activity. When cyclohexane monocarboxylate is incubated with the oxidized enzyme at pH 5.1 in the presence and absence of added (NH&S04, the acyl phosphatase is not inactivated, and the dehydrogenase activity is completely recovered when the oxidized enzyme treated in this manner is diluted in 10 mM dithiothreitol. Incubation of native glyceraldehyde 3-phosphate dehydrogenase with 25 mM 3-cyclohexene-1-carboxylate in the presence and absence of added (NH&SO4 at pH 5.1 does not inactivate the dehydrogenase activity when assayed at pH 8.5. These observations suggest that the sulfenic acid at the active site of the acyl phosphatase might add to the olefinic bond of 3-cyclohexene-lcarboxylate as described by Equation 3 at pH 5.1 in the pres-C!nCe Of 0.1 M (NH.&%.
To investigate the curious requirement of (NH&SO, for the inactivation of the acyl phosphatase by 3-cyclohexene-l-carboxylate further, the oxidized enzyme (3.5 mg per ml) was incubated with varying concentrations of different salts for 15 min at room temperature in the presence of 12.5 mM 3-cyclohexene-l-carboxylate at pH 5.0 at which time samples of the reaction mixtures were assayed. Of the salts tested over the concentration range of 0 to 200 mM, (NH&S04 was the most effective in promoting the inactivation by this olefin. For instance under these reaction conditions, 70% inactivation of the acyl phosphatase was observed in the presence of 0.15 M (NH4 $04, while the addition of 0.15 M Na2S04, 0.15 M NH&l, and 0.15 M NaCl resulted in 58, 20, and S70 inactivation of the acyl phosphatase, respectively. Fig. 7  room temperature at pH 7.5. Addition of dithiothreitol to the inactivated acyl phosphatase did not reactivate the dehydrogenase activity, indicating that a covalent modification of the enzyme had occurred. Dihydropyran has been shown to react with sulfenyl compounds to form substituted thioethers (8). When the reduced form of glyceraldehyde a-phosphate dehydrogenase is treated with dihydropyran under the same conditions, the dehydrogenase activity is not inactivated. Fig. S shows that 30 mM tetrahydrophthalimide completely in-activates the acyl phosphatase activity catalyzed by the sulfenic acid form of glyceraldehyde 3-phosphate dehydrogenase at pH 6.0 at room temperature within an hour. The addition of acetyl phosphate to a reaction mixture slightly enhanced the rate of inactivation of the acyl phosphatase by tetrahydrophthalimide. The addition of dithiothreitol to the inactivated acyl phosphatase did not lead to the recovery of the dehydrogenase activity, again indicating that a covalent modification of the enzyme had occurred.
Tetrahydrophthalimide does not inactivate the dehydrogenase reaction catalyzed by reduced glyceraldehyde 3-phosphate dehydrogenase when added to it directly.

Isolation and Sequence Characterization of Tryptic Peptide Which
Contains Cys-149 after Inactivation of Acyl Phosphatase Activity by Tetrahydrophthalimide-The amino acid sequence of pig muscle glyceraldehyde 3-phosphate dehydrogenase has been determined by Harris and Perham (18). Each of the identical subunits con tains 3 cysteine residues in addition to Cys149, the catalytically active sulfhydryl group. One of these, (:ys-153, is 4 residues removed from Cys-149, and in trypt,ic digests of the enzyme, resides on the same peptide as Cys-149.
Advantage was taken of this convenient feature to isolate specifically this tryptic peptide after the acyl phosphatase activity catalyzed by the sulfenic acid form of glyceraldehyde Y-phosphate dehydrogenase was inactivated irreversibly with tetrahydrophthalimide.
The experimental procedure for this isolation is described below.
Oxidized glyccraldehyde X-phosphate dehydrogenase (170 mg or 4.7 pmoles of enzyme subunit) was incubated with 33 mM tetrahydrophthalimide for 150 min in 14.8 ml of 0.1 M sodium succinate, pH 6.0, at which time 93% of the acyl phosphatase activity was inactivated.
The pH of the reaction mixture was then adjusted to 7.8 by the slow addition of 1 M Tris base. Then 75 pmoles of [1-r4C]iodoacetate with a specific radioactivit,y of 2.2 X 10" cpm per pmole and solid urea to a final concentration of 8 M were added. The reaction mixture was warmed to room temperature and was incubated for 2 hours to carboxymcthylate the available sulfhydryl groups with [l-"(:]iodoacetatr. Urea and excess reagents were then removed from the protein by exhaustive dialysis against I mM HCl.
Trypsin, 1.7 mg, and solid NH&HCOa to a final concentration of 0.5% were then added to the dialyzed protein.
The resulting solution was then stirred slowly at 37" for 6 hours at which time the tryptic digest was freeze-dried.
The freeze-dried digest contained 13.5 pg atom of r4C. If each of the four sulfhydryl groups per subunit had been alkylated with [I-i4C]iodoacetate, it would have contained 18.8 pg atoms of r4C.
The radioactive peptides in the tryptic digest were partially purified by successive gel filtration steps on columns (3.0 x 100 cm) of Sephadex G-25 and Sephadcx G-50, using 0.05 M NH40H as solvent.
All of the radioactivity after the Sephadex G-25 step appeared in the void volume.
The Sephadex G-50 gel filtration step resolved the radioactive material into two peaks. Peak Sl which appeared near the void volume contained 4.5 pg atoms of "C and contains the [1-r4C]carboxymethyl derivative of Cys-281 which resides on a tryptic peptide which contains residues 269 to 306 (18).
Peak SII, the radioactive material which was retained on the Sephadex G-50 column was freeze-dried and contained 8.6 pg atoms of r4C. It was subjected to anion exchange chromatography with a pH gradient as described in the legend of Fig. 9. The elution profile of the DEAE-Sephadex fractionation of the peptides in Peak SII is shown in Fig. 9. The radioactive peaks labeled I, ZZ, and ZZZ were pooled separately, freeze-dried, and desalted on a column ( acyl phosphatase which were retained by Hephadex (i-50. Peak 811 from the G-50 purification step described in the text was dissolved in 5 ml of 10/o collidine, 1% pyridine which was adjusted to pH 8.0 with acetic acid and was applied to a column (2.0 X 25 cm) of I>EAE-Senhadex A25 which was eauilibrated with the same buffer. Th;? column was washed with 100 ml of the starting buffer and was then eluted with a pH gradient which employed a mixing chamber which contained 600 ml of the starting buffer and a reservoir chamber which contained BOO ml of 1% collidine, 1% pyridine adjusted to pH 6.5 with acetic acid. 0, optical density at 750 nm of 100.~1 samples of 5.0.ml fraction treated with the Folin reagent (14); 0, counts per min of 14C in 100-J samples of the 5.0.ml collected fractions.
After desalting, l'eak 1 contained 2.2 pg atoms of r4C; Peak II contained 3.5 g atoms of 14('; and Peak III contained 0.8 g atom of Y'.
Each of the radioactive peptidcs in l'raks 1, 1 I, and 111 was then purified to homogeneity by succcssivc high voltage clcctrophoresis steps at pH 6.5 and 8.9. After clution from paper, samples of each of the radioactive pcptidcs wcrc subjected to amino acid analysis.
These analyses revealed that the radioactive pnptide purified from Peak I1 contained the [r4<']carboxymethyl derivative of Cys-244 which resides on a tryptic peptide that contains residues 232 to 245 (18) and that the radioactive peptide purified from I'cak I II contained the [r4C!]carboxymethylcysteine sulfoxide derivative of the same pcptide.
It has been shown that the atmospheric oxidation of carboxymcthylcysteine in peptides to the sulfoxide changes the ion exchange behavior of the peptides (16).
The amino acid composition of the [W]carboxymcthylated peptide purified from the DEAF-Sephadex I'eak I, which was surprisingly acidic on electrophoresis at pH 6.5, is shown in Table  111. It is clear that this peptide is a derivative of the tryptic peptide of pig glyceraldehyde 3.phosphate dehydrogenase which contains Cys-149. The amino acid analysis of this peptide showed that it contained approximately 1 residue of carboxymethylcysteine and approximately 1 residue of cysteic acid and no new ninhydrin-positive peak on the chromatogram that might be derived from a tetrahydrophthalimide derivative of Cys-149. Since cysteic acid appears near the void volume of the chromatographic system employed, the possibility that the apparent cysteic acid peak might represent the P-hydroxythioether derivative of Cys-149 with two free cis-carboxyl groups as shown by Structure 1 was considered.
To investigate this possibility, 0.6 pmole of the peptide (407, of the 1.5 pmoles isolated) were hy drolyzed with 6 M HCI at 110" in an evacuated sealed tube for 30 hours. The freeze-dried hydrolysate was then applied to as a The preparation of the radioautograph which superimposes a guide strip stained with ninhydrin is described in the text.
15-cm band on Whatman No. 3MM paper and was subjected to high voltage paper electrophoresis at pH 1.9 at 3 kv for 45 min. Fig. 10 shows a radioautogram which is superimposed over a ninhydrin-collidine (17).stained guide strip which includes two mixtures of marker amino acids. Qualitatively, Fig. 10 indicates that the peptide contains 1 residue of lysine, 1 residue of carboxymethylcysteine which is radioactive, and 1 residue of cysteic acid. The material which migrated with cysteic acid at pH 1.9 had the same electrophoretic mobility as authentic cysteic acid when subjected to electrophoresis at pH 3.5, 6.5, and 8.9,. It also had the same RF as authentic cysteic acid when subjected to paper chromatography using butanol-pyridine-acetic acid-water (150: 100: 30:120) as solvent.
Therefore, we conclude that the acid hydrolysis at 110" of the tryptic peptide which contains Cys-149 which was isolated after inactivating the acyl phosphatase activ-ity with tetrahydrophthalimide, and then carboxymethylating with [VCliodoacetate gives rise to 1 residue of [14C]carboxymethylcysteine and 1 residue of cysteic acid.
The cysteic acid identified in the acid hydrolysates described above may have been formed directly when the acyl phosphatase was inactivated with tetrahydrophthalimide or it may have been formed during acid hydrolysis of an adduct of Cys-149 in the tryptic isolated peptide by 6 M HCl at 110". To distinguish between these possibilities, the remainder of the radioactive peptide, 0.9 pmole, was digested with 50 pg of pronase for 18 hours at 37" in 0.5 ml of 0.5% NH4HC03.
The pronase digest was subjected to paper electrophoresis at pH 6.5 for 50 min at pH 6.5 at 3 kv. The ionogram was radioautographed and a guide strip was stained with the ninhydrin-cadmium reagent (17). Inspection revealed, in addition to a high concentration of neutral material, a single basic dipeptide which was Ala-Lys and a number of radioactive acidic peptides which were ninhydrin-positive. In addition there was a single acidic ninhydrin-positive peptide that was not radioactive that had a mobility relative to aspartic acid of 0.55. This peptide was eluted and 5% of it was hydrolyzed with 6 M HCl at 110" for 24 hours in an evacuated sealed tube. Amino acid analysis of the dried hydrolysate revealed that it contained 3.9 nmoles of cysteic acid and 4.6 nmoles of threonine and traces of serine, glutamic acid, and glycine.
The remainder of the peptide was separated into three aliquots which were freeze-dried.
One of the aliquots was hydrolyzed in an evacuated sealed tube for 24 hours at 110" with 300 ~1 of 6 M HCl. A second aliquot was hydrolyzed with 300 ~1 of 6 M HCl at 37" for 30 hours in an evacuated sealed tube. The third aliquot was untreated.
After freeze-drying, the three aliquots treated as described above were then applied, along a common origin, on Whatman No. 3MM and subjected to electrophoresis at pH 6.5 for 45 min at 3 kv. The ionogram was stained with the ninhydrincadmium reagent (17). A drawing which represents the relative intensity of ninhydrin-positive material observed on this ionogram is shown in Fig. 11. The aliquot of the dipeptide which was hydrolyzed at 110" with 6 M HCl revealed cysteic acid and a neutral amino acid which is threonine.
The partial acid hydrolysate revealed some neutral ninhydrin-positive material, and a new ninhydrin-positive spot with a mobility slightly greater than that of glutamic acid and a very faint ninhydrin-positive spot which migrated with the same mobility as the untreated peptide.
Based on Cfford plots (22) in which the log of the electrophoretic mobility relative to aspartic acid is plotted against the log of the molecular weight of the peptide, we propose that the untreated pronase peptide has the composition shown in Structure II. Also, on the basis of an Offord plot, the ninhydrin-positive material in the partial acid hydrolysate which migrated slightly ahead of glutamic acid is the derivative shown by Structure I. The untreated peptide which is converted to cysteic acid and threonine by acid hydrolysis at 110" for 24 hours, falls on an Offord plot with a net negative charge of -1 if the molecular weight of the compound is based on Structure II. Similarly the ninhydrin-positive material with a mobility slightly greater than that of glutamic acid falls on an Offord plot with a net negative charge of -2 if the molecular weight of the compound is based on Structure I.
These results suggest that tetrahydrophthalimide forms an adduct with Cys-149 when it inactivates the acyl phosphatase activity catalyzed by the sulfenic acid form of glyceraldehyde 3-phosphate dehydrogenase and that the adduct is converted to cysteic acid on acid hydrolysis with 6 M HCl at 110". We suggest that the radioactive tryptic peptide isolated from Peak I of the DEAE-Sephadex column has the sequence and structure shown below.
;" s I Ile-Val-Ser-Asn-Ala-Ser-Cys-Thr-Thr-R' I Asn-Cys-Leu-Ala-Pro-Leu-Ala-Lys where R is Structure III below and R' is -CH214C00-. The monoamide shown in Structure III accounts for the acidity of this peptide when it was subjected to electrophoresis at pH 6.5 and probably arose by hydroxysis of the P-hydroxythioether adduct of tetrahydrophthalimide with Cys-149 during the Sephadex This solvent would be sufficiently basic to hydrolyze the imide to the monoamide. DISCUSSION The experimental results presented clearly indicate that ['"Cldimedone reacts covalently with sulfenic acid form of glyceraldehyde 3-phosphate dehydrogenase to form a covalent derivative of Cys-149.
This inactivates the acyl phosphatase reaction catalyzed by the oxidized form of the enzyme. Since the covalent label is not removed by dithiothreitol, we suggest that the C-2 carbanion tautomer of dimedone reacts with the sulfenic acid at the active site of the enzyme as shown in Equation 3. If an enolate tautomer of dimedone had reacted with the sulfenic acid in a nucleophilic displacement reaction to form a sulfenyl ester, the label would have been removed by dithiothreitol and the dehydrogenase activity catalyzed by reduced glyceraldehyde 3phosphate dehydrogenase would have been reactivated.
The results presented also show that the olefins 3-cyclohexene-1-carboxylate, dihydropyran, and tetrahydrophthalimide inactivate the acyl phosphatase reaction catalyzed by the sulfenic acid form of glyceraldehyde 3-phosphate dehydrogenase.
Since the addition of dithiothreitol to the acyl phosphatase inactivated with t'he olefins does not reactivate the dehydrogenase activity catalyzed by reduced glyceraldehyde 3-phosphate dehydrogenase, and since sulfenyl compounds in general (7,8) form adducts with olefins, we suggest that the sulfenic acid at the active site of the acyl phosphatase forms addition products with the olefins during the activations described.
Sequence analysis of the tryptic peptide which contains Cys-149 after the acyl phosphatase was inactivated with tetrahydrophthalimide strongly supports this hypothesis.
There are three other enzyme-catalyzed reactions that have been reported to be inactivated irreversibly by dimedone.
Under certain conditions dimedone inactivates papain as reported by Morihara (23). Morihara showed that papain prepared by the method of Kimmel and Smith (24) which was inactive in the presence of reducing agents was inactivated when it was incubated with dimedone before the addition of activating t.hiols. Since the sulfhydryl group at the active site of papain (Cys-25) can be oxidized to a sulfenic acid by Hz02 (25) and that evidence has been presented that native, inactive papain is in part in the sulfenic acid form (26), it is possible that dimedone prevents the activation of Morihara's preparation of papain by reacting covalently with the sulfenic acid at the active site to form a covalent derivative of Cys-25.
Zeller (27)  III amine oxidase isolated from beef plasma, an inactivation which we have recently repeated in our laboratory with the highly purified (specific activity >500) beef plasma non-flavin amine oxidase prepared by the method of Yasunobu and Smith (28). It has been postulated for years that the non-flavin amine oxidases contain pyridoxal phosphate or a related prosthetic group. This is based primarily on the fact that these amine oxidases are inactivated by carbonyl reagents.
However, conclusive evidence that such a prosthetic group which contains an aldehyde moiety has never been presented (29). We have recently presented evidence that the sulfenic acid form of glyceraldehyde Y-phosphate dehydrogenase possesses a limited amine oxidase activity (30). Gorkin and his associates have shown that the oxidation of a sulfhydryl group in flavin amine oxidase converts the enzyme to an amine oxidase t,hat is sensitive to carbonyl reagents and is specific for primary amines (31). These are the properties of the non-fiavin amine oxidases. It is therefore tempt,ing to suggest the following reaction mechanism for the non-flavin amine oxidases which hypothesizes a sulfenic acid and a sulfenamide intermediate on the reaction pathway as described by Equations 5 to 8.
ESH + 02 -ESOH + H202 This reaction scheme is consistent with the ping-pang steady state kinetics exhibited by the non-flavin amine oxidase from pig plasma which suggests that two covalently modified forms of the enzyme are produced in the reaction sequence (32). The reaction scheme is also consistent with the products of the non-flavin amine oxidase which are H202, NH,, and the aldehyde (27). The non-flavin amine oxidases contain Cur1 which is necessary for activity (33). It is interesting that Cu" in the presence of O2 oxidizes Cys-149 in glyceraldehyde 3-phosphate dehydrogenase to a sulfenic acid and that Cu" in the presence of O2 oxidizes a sulfhydryl group in the mitochondrial amine oxidase to a sulfenic acid and converts the amine oxidase to one that is specific for primary amines and is 110 longer inactivated by pargyline (31), a flavin-specific reagent.
Dimedone has also been reported to inhibit the oxidation of NADH by the mitochondrial electron transport chain by Weiss (34). Skidmore and Whitehouse have reported that dimedone and dimedone derivatives decrease the P: 0 ratio when rat liver mitochondria are t#reated with t)he compounds (35). Recent experiments performed in this laboratory have shown that the incubation of submitochondrial particles from beef heart with high phosphorylating activity prepared by the method of Hansen and Smith (36) with 7.5 mM dimedone for 2 min at 38" inhibits ATP-driven reversed electron flow at coupling site I by 50%. Reversed electron flow was determined by the reduction of NAD+ by succinate which requires ATP as an energy source as described by Sanadi and his colleagues (37). Since we have postulated a scheme for chemical coupling in oxidative phosphorylation which 6243 postulat,es a sulfenyl carboxylat,e as a nonphosphorylated high energy int,ermediat,e (3)) we are pursuing the possibility that these fairly high concentrations of dimedone are reacting covalently with a protein sulfenic acid or sulfenic acid derivative in the mitochondrion and are thus inhibiting oxidative phosphorylation.