Isolation and Characterization of a Homogeneous lsoenzyme of ...

ARCHIVES OF BIOCHEMISTRY AND BIOPHYSICS

Vol. 288, No. 2, August 1, pp. 621-633, 1991

Isolation and Characterization of a Homogeneous lsoenzyme of Wheat Germ Acid Phosphatase

Parvin P. Waymack and Robert L. Van Ettenl Department of Chemistry, Purdue University, West Lafayette, Indiana 47907-1393

Received November 27, 1990, and in revised form April 17, 1991

An acid phosphatase (orthophosphoric monoester phosphohydrolase, acid optimum; EC 3.1.3.2) isoenzyme from wheat germ was purified 7000-fold to homogeneity. The effect of wheat germ sources and their relationship to the isoenzyme content and purification behavior of acid phosphatases was investigated. Extensive information about the purification and stabilization of the enzyme is provided. The instability of isoenzymes in the latter stages of purification appeared to be the result of surface inactivation together with a sensitivity to dilution that could be partially offset by addition of Triton X- 100 during chromatographic procedures. Added sulfhydryl protecting reagents had no effect on activity or stability, which was greatest in the pH range 4-7. The purified isoenzyme was homogeneous by polyacrylamide gel electrophoresis and exhibited the highest specific activity and turnover number reported for any acid phosphatase. The molecular weights of the pure isoenzyme and of related isoenzymes from wheat germ were found to be identical (58,000). The pure isoenzyme contained a single polypeptide chain and had a negligible carbohydrate content. The amino acid composition was determined. Of the various reasons that were considered to explain isoenzyme occurrence, a genetic basis was considered most likely. The enzyme was found to exhibit substrate inhibition with some substrates below pH 6, while above pH 8 it exhibited downwardly curving Lineweaver-Burk plots of the type that are generally described as “substrate activation.” The observation of a phosphotransferase activity was consistent with the formation of a covalent phosphoenzyme intermediate, while inactivation by diethyl pyrocarbonate was consistent with the presence of an active site histidine. 0 1991 Academic Press, Inc.

The ubiquitous distribution and general characteristics of acid phosphatase (orthophosphoric monoesterphos-

1 To whom correspondence should be addressed.

0003.9861/91$3.00 Copyright 0 1991 by Academic Press, Inc. All rights of reproduction in any form reserved.

phohydrolase, acid optimum; EC 3.1.3.2) have been described in extensive reviews (1, 2). Most studies of acid phosphatase, particularly those from plant sources, are largely of a descriptive nature. With a few exceptions (no- tably, human prostatic acid phosphatase) the acid phosphatases occur in very small quantities, are unstable in dilute solution, and are subject to surface denaturation when purified. These factors, together with a tendency to occur in multiple forms or as isoenzymes, makes the isolation of highly purified acid phosphatases difficult.

Studies of acid phosphatases from various sources suggest a role in phosphate mobilization for the enzyme (3- 5). The lysosomal location of the high molecular weight mammalian liver acid phosphatases (6) can be compared to the location of seedling acid phosphatase activity in sphereosomes, which have been described as analogous to lysosomes (7).

Wheat germ acid phosphatase (WGAP)’ exists in multiple chromatographic and electrophoretic forms which have been called isoenzymes. Chromatographic behavior on DEAE-cellulose under varied conditions have shown multiple peaks of activity (8-10). Multiple electrophoretic forms have been found by starch and polyacrylamide gel electrophoresis (10, 11).

Acid phosphatase zymograms of extracts from wheat plant tops appear to contain a distinct isoenzyme under field conditions of phosphorus deficiency (12). Changes of isoenzyme patterns in the various stages of germination of hard red wheat have been studied (5). Polyacrylamide gel electrophoresis and activity staining were used to de- tect changes in the number of multiple forms and chang- ing mobilities in the dormant seed, germinated seed, co- leoptile, and first leaf tissue. In seed, there were five bands which contained most of the activity in the more mobile bands. In contrast, the first leaves contained four bands with most of the activity distributed among the less mobile

* Abbreviations used: WGAP, wheat germ acid phosphatase; DEP, diethyl pyrocarbonate; SDS, sodium dodecyl sulfate.

621

622 WAYMACK AND VAN ETTEN

bands. It was proposed that the combination of distinct subunits controlled by different genes could give rise to hybrid enzymes with different electrophoretic mobilities (5).

The observation of increased acid phosphatase activity during germination, and of greater isoenzyme mobility as germination progressed, suggests that the possible synthesis, activation, or conversion to new forms of the enzyme occurs during germination. However studies of seed germination under conditions designed to preclude de nouo synthesis of active enzyme showed the typical increases in acid phosphatase production (13). It was concluded that the acid phosphatase activity arose by activation of preformed macromolecules, i.e., zymogens. These obser- vations suggest that the interconversion of isoenzymes may occur.

Interchangeable subunit structures, if present, could explain some of the multiple electrophoretic forms of wheat acid phosphatase. Studies of the acid phosphatase of Drosophilia melanogaster showed the presence of AA, BB, and AB electrophoretic variants (14). Sunflower seed acid phosphatase has been shown to consist of two subunits (15). Because of the existence of acid phosphatases with molecular sizes as low as 18 kDa (16), it was conceivable that two or even four subunits could be present in wheat germ acid phosphatase.

The interaction of subunits is the primary mechanism which has been implicated in the property of negative cooperativity. This type of kinetic behavior has been observed at the pH optima of other acid phosphatases containing subunits, including rat liver (6), human prostate (17), and human liver at pH 7.3 (18). Additionally, rat mammary carcinoma acid phosphatase (19) can be sur- mised (20) to contain subunits. Since the present report revealed a type of negative cooperativity by wheat germ acid phosphatase at pH 7.8 and above, the studies of subunit structure and enzyme purity were of additional interest.

In characterizing wheat germ acid phosphatase, some uncertainty exists as to what constitutes a nonspecific acid phosphatase. This is a somewhat loosely defined enzyme classification based on an acid pH optimum for phosphomonoesterase hydrolysis. While acid phosphatases typically have broad substrate specificity, the reported polyphosphatase and phosphodiesterase activities of many such enzymes raise questions of purity of the preparations involved. One frequently cited work (8) reported confirmation of an earlier report (21) of the presence of iron in four yellow isoenzymes which also hydro- lyzed carboxylesterase substrates.

These questions of enzyme structure, substrate specificity, and unusual kinetic behavior are best answered with studies using the highest purity enzyme. Therefore, purification to homogeneity was undertaken to allow reliable characterization of wheat germ acid phosphatase and to

clarify earlier reports of its properties. The present work describes in detail the purification to homogeneity of an isoenzyme of wheat germ acid phosphatase, as well as a characterization of some of its most interesting physical and kinetic properties.

EXPERIMENTAL PROCEDURES

Purification of WGAP All steps were carried out at 4°C except the initial acetone precipi-

tation, which was performed in a fume hood at room temperature.

Step 1. Initial extraction. Dry wheat germ was finely ground in a Waring blender and extracted with 0.3 M pH 4.0 acetate buffer (10 ml per gram of wheat germ) by stirring at lo-min intervals for 1 h.

Step 2. Acetone precipitation. At timed intervals, cold acetone (-ZO’C) was slowly added with stirring to batches of the extract from Step 1, to give a final concentration of 55% (v/v). After resuspending the initial precipitate several times with gentle swirling and shaking, the precipitate cleanly settled after 30-60 min, leaving a clear supernatant that was decanted and discarded. The remaining solution containing the precipitate was centrifuged at 159Og for 5 min. The precipitates from several successive batches were collected in the same set of centrifuge bottles. The final centrifugation was extended to 15 min at 6370g to minimize acetone carry-over in the pellet and to allow rinsing out of acetone with cold water. The largely insoluble pellet was suspended by gentle mechanical stirring in a volume of cold distilled water equal to about one-fiftieth the final Step 1 volume. If the volume was adequate, centrifugation of the suspension at 13,300g for 20 min yielded a clear amber supernatant. (A clear supernatant indicates the volume used to dissolve the pellet was adequate.)

Step 3. Ammonium sulfate precipitation and acetone remoual. To the clear supernatant fraction from Step 2 was added 432 g ammonium sulfate/liter. It was convenient to use the nomogram of Dixon (22) using the 65% saturation point although the procedure was carried out at 4°C. The mixture was allowed to stand for 30 min and then centrifuged at 4440g for 10 min. Acetone was discarded with the supernatant fraction. After rinsing with cold water, the pellet was dissolved in a volume of 0.3 M pH 4.0 acetate buffer equal to one-half of the final Step 2 volume and dialyzed against the same buffer for 8 h with several changes. The dialysate was centrifuged at 11,300g for 15 min.

Step 4. Ammonium sulfate fractionation. The dialysate from Step 3 was diluted to 8-10 mg/ml (protein assay) with 0.3 M acetate buffer, pH 4.0, and ammonium sulfate (210 g/liter) was added. The ammonium sulfate solution was allowed to stand for 30 min and centrifuged at 11,300g for 15 min. The pellets were discarded and the clear supernatant fractions pooled. Ammonium sulfate (91 g/liter) was added to the supernatant and after 30 min the solution was centrifuged as above. The centrifuge bottle walls and pellet surface were rinsed with cold water to remove supernatant ammonium sulfate solution which would interfere with the next step. The pellet was dissolved in cold water, assayed for protein concentration, and diluted to 4-6 mg/ml (The protein concentration is critical for Step 5).

Step 5. Methanol precipitation. The concentration of methanol was brought to 23% (v/v) by slow addition with stirring of 0.3 ml cold (-2O’C) methanol per milliliter of Step 4 material. A precipitate formed during centrifugation at 15,960g for 60 min. The largely insoluble pellet was dissolved in the starting Step 5 volume of cold water and centrifuged at 48,000g for 45 min. The supernatant was used for the next step.

Step 6. G-75 Sephadex gel filtration. The Step 5 fraction was concentrated and equilibrated with 0.01 M acetate buffer, pH 5.2, with 0.1 M NaCl and 0.1% Triton X-100 added (G-75 buffer) by ultrafiltration with an Amicon PM-10 membrane. Alternatively it was precipitated by 55% acetone as previously described and then dissolved in G-75 buffer. The sample was cleared by centrifugation at 48,OOOg for 20 min prior

HOMOGENEOUS ISOENZYME OF WHEAT GERM ACID PHOSPHATASE 623

to application to a 5 X 100 cm G-75 Sephadex column equilibrated with G-75 buffer. Separation at a flow rate of 35 ml/h gave a single activity peak which corresponded to a shoulder on the trailing edge of a large 280-nm absorbance peak. Fractions containing two-thirds of the activity were combined and concentrated by ultrafiltration to enhance stability.

Step 7. SP-Sephadex chromatography. The concentrate from Step 6 was applied to a 1.6 X 15 column of SP-Sephadex equilibrated in G- 75 buffer. The column was eluted at 12 ml/h with a linear NaCl gradient. The G-75 buffer was in the mixing chamber and G-75 buffer containing 0.5 M NaCl in the reservoir. The fractions were monitored for protein by a modified Folin method (23). The elution profile had two main activity peaks following a small void volume activity peak (~1% of total activity). The center of the first peak (-9% of total activity) was eluted at 0.14 M NaCl while the second peak (-90% of total activity) was centered at 0.23 M NaCl. The protein assay indicated a correspondence of activity and protein in the main WGAP peak.

Step 8. DEAE-cellulose chromatography. The combined, concentrated (Amicon PM-lo) main peak WGAP fractions from Step 7 were dialyzed against 6 liters of 5 mM, pH 7.4, Tris-HCl containing 0.1% Triton X-100 (pH 7.4 buffer) with two buffer changes during 9 h of dialysis. In order to minimize nonspecific protein adsorption, DEAE- cellulose was treated by successively washing with 1 M HCl, 1 M NaOH, pH 7.4, buffer, 1 mg/ml bovine serum, and pH 7.4 buffer containing 1 M NaCl before final equilibration with pH 7.4 buffer. The enzyme sample was applied to a 0.9 X 10 cm column and washed with 25 ml pH 7.4 buffer. The enzyme was eluted at 15 ml/h with a linear NaCl gradient generated with 75 ml of pH 7.4 buffer containing 0.3 M NaCl and 0.1% Triton X-100 in the reservoir and 75 ml of pH 7.4 buffer in the mixing chamber. The elution profile showed a protein peak corresponding to the single activity peak. The fractions with the highest specific activity were combined, the pH adjusted to 5.2 with 0.5 M pH 4.0 acetate buffer and concentrated by ultrafiltration. The steep gradient and rapid flow rate of this column are designed to minimize the length of time the enzyme remains at high pH and to minimize dilution of protein, since both factors reduce enzyme stability.

Step 9. Rechromatogrophy on SP-Sephadex. The concentrated fractions from Step 8 were dialyzed against the pH 5.2 G-75 buffer. The sample was applied to a 0.9 X 10 cm column as in Step 7 and rinsed with 25 ml of G-75 buffer. The column was eluted at 8 ml per hour by a linear NaCl gradient with 50 ml G-75 buffer in the mixing flask and 50 ml of G-75 buffer containing 0.5 M NaCl in the reservoir. The fractions with highest specific activity were concentrated as soon as possible.

Specific Activity Assay The standard condition for measuring activity was assay in 0.1 M

acetate, pH 4.6, using 5 mM p-nitrophenyl phosphate at 25’C. In a typical assay, 2.00 ml was incubated 2-5 min and quenched with 1.0 ml of 0.4 M NaOH and the 400-nm absorbance was read versus a blank in which NaOH was added before enzyme. One unit (U) of enzyme is that required to release 1 pmol p-nitrophenol per min.

Molecular Weight Measurements A 2.5 X 100~cm column of Sephadex G-200 was calibrated with 0.05

M pH 7.0 phosphate buffer using blue dextran and standard proteins (catalase, aldolase, bovine serum albumin, soybean trypsin inhibitor and cytochrome c) applied in a volume of 1.0 ml. A linear least squares program was used to determine the linear relationship between log molecular weight and elution volume. Sodium dodecyl sulfate-gel electrophoresis for polypeptide molecular weight was conducted under dena- turing and reducing conditions (24). Electrophoresis was performed at 8 mA per gel for 8 hr followed by Coomassie blue staining. Mobilities were calculated as recommended by Weber and Osborne (24) and were the average of at least two determinations. Samples containing IO-43 pg WGAP at a specific activity of 433 unit/mg were applied to each gel.

Carbohydrate Determination

A phenol-sulfuric acid method (26) was used for carbohydrate determination in the early stages of purification. A more sensitive method (27), verified by ovalbumin and lectin from Bandeiruea simpZic$oZia, was used in the latter stages of purification.

Sialic Acid Assay and Neuraminidase Treatment

The mixture of WGAP isoenzymes found in the Step 4 ammonium sulfate fraction was treated with neuraminidase from Clostridium per- fringens (l-4 units/ml) at pH 5.0 in acetate buffer at 25’C. Aliquots were withdrawn after incubation for 24, 48, and 72 hr, electrophoresis was performed at pH 4.4, and the gels were stained for acid phosphatase activity. Electrophoretically homogeneous isoenzymes were similarly treated.

Carbohydrate Detection after Polyacrylamide Gel Electrophoresis

Following polyacrylamide gel electrophoresis, localization of carbohydrate was attempted using a sensitive fluorescence method (25) with glycoprotein standards which allowed the detection of as little as 40 ng of carbohydrate under uv illumination.

Amino Acid Composition

Homogeneous WGAP (605 U/mg) was absorbed on SP Sephadex C- 50 at pH 5.2, washed to remove Triton X-100, and eluted with 0.1 M pH 7.0 ammonium carbonate. Samples (0.2 mg) were desalted on G-25 Sephadex and analyzed according to Ref. (15).

Enzyme Stability

The stability of WGAP at the concentration and conditions employed in the continuous rate assays at high pH was investigated using the same enzyme solution and concentration throughout the pH range 2.45 to 10.0. To 2.40 ml of each buffer at 25.0°C, 50 ~1 of solution containing 0.8 units of WGAP was added. Aliquots of 50 ~1 were withdrawn and analyzed by the standard activity assay. The half-life of the enzyme was estimated from the linear portion of log activity versus time. The storage stability of partially purified WGAP (Step 5, methanol precipitation) was also monitored as a function of pH at 4°C for 6 months. Samples were incubated at 4.0 mg/ml in 0.025 M buffers as follows: glycine, pH 2.93; acetate, 3.9 and 5.0; maleate, 5.95; Tris, 7.4; barbital, 8.0 and 9.0.

Metal Ion Effects

The effect of 0.1 to 10 mM added metal ions upon activity in 0.1 M

pH 5.0 acetate was studied. Concentratedp-nitrophenyl phosphate was added to 2.00 ml of the metal ion solutions to give a final concentration of 5.0 mMp-nitrophenyl phosphate. After 2.0 min incubation at 25.O”C, the reaction was quenched and read at 400 nm, after removal of any precipitates by centrifugation.

The effect of preincubation of the enzyme with metal ions in 0.1 M

pH 5.0 acetate was studied by addition of 0.050 ml of WGAP containing 175 U/ml (218 U/mg) to 1.0 ml of 1.05 mM metal ion solution at 25°C. Aliquots of the mixture were withdrawn and assayed along with a control experiment with no metal ions. Control experiments showed that the effect of metal ions carried over into the assay mixture was negligible.

The effects of two chelating agents upon activity were investigated. At pH 5.0, 0.1 M acetate contained 5 mM p-nitrophenyl phosphate and either 10 mM EDTA or 5 mM o-phenanthroline. At pH 7.0,0.03 M barbital containing 0.1 M NaCl was used with 10 mM EDTA or 5 mM o-phenanthroline added. Assay was as above, using control experiments without chelators. The effects of added EDTA upon the stimulatory effect of 1.0


mM Mg+‘, 1.0 mM Co+*, and 0.05 mM NazMoO, were investigated with the above conditions.

The effect of preincubation in 25 mM EDTA and 5 mM o-phenanthroline was studied in pH 5.0, 0.1 M acetate at 4 and 25“C. To 0.50 ml of these mixtures (and controls without chelating agents), 0.050 ml of WGAP containing 216 U/ml (308 U/mg) was added and aliquots were withdrawn at 24-hr intervals for assay.

Steady-State Kinetics

Substrate solutions were prepared by dilution of concentrated solutions with the respective buffers. The concentration of p-nitrophenyl phosphate was varied from 0.1 to 10 mM, using 0.05 M acetate, pH 5.0, with 0.1 M NaCl added. At pH 6.0, 0.05 M 3,3-dimethylglutarate buffer containing 0.1 M NaCl was used. To 2.00 ml of substrate solutions at 25’C, 25-50 PL of WGAP was added and quenched with 0.4 or 1.0 ml of 1.0 M NaOH. At pH 7.0 and above, direct continuous assays were employed (25-50 ~1 WGAP added to 2.40 ml of substrate thermostatted at 25°C). At pH 7.0, 0.05 M 3,3-dimethylglutarate buffer containing 0.1 M NaCl and 0.05 to 20.0 mM was used. pH values 7.0, 7.5 (0.05 M 3,3-dimethylglutarate buffer with 0.1 M NaCl), 8.0,8.5, and 9.0 (0.05 M barbital with 0.1 M NaCl) were used with the range of 0.1 to 10 mM substrate.

Heat Treatment and Specificity

Partially purified WGAP (Step 7) having a specific activity of 308 mg/dl was inactivated by heating 0.2 ml samples (in pH 5.2 0.1 M acetate)

for 2 min at water bath temperatures from 35 to 75°C and then rapidly cooling them (ice water). Activity remaining was assayed in duplicate using the standardpH 4.6 assay and directly at pH 8.5 in 0.05 M barbital containing 0.1 M NaCl and either 10 mM p-nitrophenyl phosphate or 10 mM (his)-p-nitrophenyl phosphate. The same buffer was used in a single determination withp-nitrophenyl phenylphosphonate as substrate.

pH Dependence of the Activation Energy

For the determination of V,,,,,., five concentrations of p-nitrophenyl phosphate were prepared by the addition of 0.10 ml substrate solutions to 2.00 ml of 0.055 M buffer containing 0.11 M NaCl. The reaction was initiated by the addition of 50 pl of a WGAP solution and quenched with 1.0 ml of 1.0 N NaOH. At low pH, the use of saturating substrate conditions (50 X K,J was used to give a good approximation to V,,,. For this, 5 mM p-nitrophenyl phosphate was used, since little or no substrate inhibition is observed at this concentration in the pH range 3-5. This facilitated testing the Arrhenius equation over a wide temperature range.

The temperature range lo-40°C was investigated at pH 5.0. The reaction in a thermostated sample was initiated by the addition of 50 ~1 of WGAP enzyme solution to 2.0 ml of 5 mM p-nitrophenyl phosphate in 0.05 M acetate buffer with 0.1 M NaCl added. The reaction was quenched by the addition of 1.0 ml of 1.0 N NaOH and the absorbance at 400 nm was determined. An identical procedure was employed with pH 3.00,0.05 M glycine at 15,25, and 37.5”C, and pH 4.0,0.05 M acetate buffer at 15.0 and 37.5”C. In each case an identically treated spectro- photometric blank was prepared in which the enzyme was added after the reaction was quenched.

Transphosphorylation

Incubation mixtures containing 50% methanol and 10 mM p-nitrophenyl phosphate were prepared by adding 0.10 ml concentrated substrate solution to 1.90 ml of buffer solutions (pH 3.5 to 9.2) and 2.0 ml of methanol. The pH of the mixture was that measured with a glass electrode. The incubation time and amount of enzyme added were adjusted to give less than 2% total hydrolysis before the reaction was quenched with 0.50 ml of 3% molybdate in 0.5 M pH 4.0 acetate buffer.

The quenched mixture was assayed for p-nitrophenol by the addition of 1.00 ml of 1 M NaOH to a 2.00-ml portion and the absorbance was measured at 400 nm. The inorganic phosphate content of the same quenched mixture was determined by the Lowry phosphate determination method. The slight effect of methanol upon the extinction coef- ficient at 700 nm (I& = 4600 M-’ cm-‘) was determined in separate experiments where equal amounts ofp-nitrophenol and inorganic phosphate were produced enzymatically in the absence of methanol. The amount of methyl phosphate ester that was produced was taken to be the difference between the moles of p-nitrophenol and inorganic phosphate produced.

Inactivation by Diethyl Pyrocarbonate Protein modification experiments with diethyl pyrocarbonate (DEP;

ethoxyformic anhydride) were conducted in 0.50 ml of 0.050 M dimethyl glutarate or barbital buffers in the pH range 6.0-8.5 to which was added 10 ~1 of WGAP solution and 25 al of an ethanol solution of DEP. So- lutions of DEP for the inactivation experiments were prepared by dilution of commercial 7.0 M DEP with absolute ethanol. Control experiments contained added ethanol without DEP. In experiments where phosphate was added as an enzyme protecting agent, NaCl was added to the phosphate-free inactivation mixture to keep the ionic strength constant. Al- iquots (25 ~1) were removed from the 25°C incubation mixture and assayed for activity for 2-4 min by the standard activity assay procedure except that 5 mM EDTA was added to allow accurate assay of molybdate- containing aliquots. The reaction was stopped by addition of 1.0 ml of 1.0 N NaOH and the absorbance at 400 nM was measured. Pseudo-first- order rate constants for the initial loss of activity were evaluated from the linear portion of plots of log activity remaining versus time. Reac- tivation with hydroxylamine was carried out by adding equal volumes of the DEP-inactivated sample and 0.2 M hydroxylamine adjusted to pH 7.0 in 0.1 M pH 7.0 dimethylglutarate (thus giving 0.1 M hydroxylamine in the reactivation mixture). The reactivation mixture was incubated at 25°C and aliquots were assayed for activity as described above.

RESULTS

The purification procedure summarized in Table I provided a reliable method for obtaining homogeneous WGAP. It was reproducible with every lot of wheat germ from soft red wheat cultivar Arthur (supplied by Acme- Evans, Indianapolis, IN). The high specific activity going into the column chromatographic steps increased the problem of activity loss from dilution and nonspecific adsorption. The addition of 0.1% Triton X-100 helped stabilize activity during G-75 Sephadex chromatography and subsequent steps. The best results were obtained when several preparations were pooled before the chromatographic steps, thus maximizing protein concentration during chromatography. Pretreatment of the DEAE-cellulose column and the steep gradient were necessary to avoid extensive activity loss. The procedure resulted in an at least 7000-fold purification of WGAP with a final specific activity of 605 U/mg when assayed at pH 4.6 in 0.11 M sodium acetate buffer with 5 mM p-nitrophenyl phosphate as a substrate. WGAP appeared homogeneous by polyacrylamide gel electrophoresis at pH 8.0 and 4.4 and gave a single band upon SDS gel electrophoresis. For two chromatographically distinct isoenzymes (the two activity peaks in Fig. 1) average molecular weights of


60,000 and 58,000 were calculated for the native proteins on Sephadex G-200 chromatography. On the basis of such experiments, it is thought that these and other isoenzymes from wheat germ have indistinguishable molecular weights. Lower molecular weight species or higher molecular weight aggregates were not observed.

From SDS-gel electrophoresis, a plot of mobility versus log molecular weight exhibited a linear relationship from which the molecular weight calculated for WGAP was 56,000 + 3000. This value is in agreement with the value determined by gel filtration, and amino acid analysis (Ta- ble II). The failure of a mixture of WGAP isoenzymes to bind to a concanavalin A-Sepharose column suggested that none of the WGAP isoenzymes contain carbohydrate chains with significant amounts of terminal mannose or galactose residues.

The inability of high concentrations of neuraminidase and long incubation times to generate new electrophoretic forms is consistent with the absence of neuraminic acid in the acid phosphatase isoenzymes. The progressive removal of carbohydrate contamination with purification (results not shown) indicated that pure WGAP contains <l% carbohydrate. This conclusion is also consistent with results obtained by the dansyl hydrazine staining procedure. A faint fluorescence was detectable only at the 43- pg WGAP protein level in gels that had been periodate- oxidized as well as in those which had not. If we assume 1% carbohydrate in WGAP, the calculated amount of carbohydrate in such a sample would be 430 ng, which is more than 10 times the minimum detection level. The sensitivity was confirmed with the standard proteins, ovalbumin, and Bandeiraea simplicifolia lectin.

The results of amino acid analysis of the isoenzyme WGAP are summarized in Table II. The residues per molecule were calculated by an integer fit method (29). A plot of the sum of the relative residuals versus total residues showed a distinct minimum at 521 residues, cor-

% 0.4

2 0.3

LT - 0.2 ::

i 0.1

IO 20 30 40 50 60 70 80 FRACTION NUMBER

FIG. 1. SP-Sephadex chromatography during Step 7 of the purification of wheat germ acid phosphatase. A 1.6 X 10 cm column was eluted with a linear NaCl gradient as described under Experimental Procedures. The activity (0) was measured by the standard assay method and the protein was measured by the Folin assay as 750 nm absorbance (0). The separate activity peaks of fractions 23-33, and 40-55 were collected.

responding to a molecular weight of 54,950 g/mol. This value (with the addition of a likely tryptophan contri- bution) agrees well with the value obtained by gel filtration and SDS-gel electrophoresis. The relative content of acidic and basic residues is consistent with the observed chromatographic and electrophoretic behavior of WGAP as a slightly acidic protein. Attempts to obtain sufficient quantities of a second homogeneous isoenzyme from wheat germ strain Arthur were unsuccessful due to the low content of any other individual isoenzyme.

Studies of the stability of WGAP at 25°C showed a dramatic loss of stability above pH 7.5 and below pH 3. Besides pH, the most serious stability problem appears to be the interrelated effects of protein concentration and surface denaturation that were evident in chromatography steps involving highly purified enzyme. In the pH range of maximal stability, pH 4 to 7, the half-life of partially purified enzyme solutions stored at 4°C at a protein con-

TABLE I

Purification of Wheat Germ Acid Phosphatase

step Volume

(ml) Protein Activity btdml) (units/ml)

Total units

Specific activity”

Recovery (%)

Purification factor

(n-fold)

1 Initial extract 64,000 10.4 0.90 57,600 0.086 (100) (1) 2 Acetone precipitation 1,320 26.1 35.3 46,600 1.35 81.0 16 3 O-65% (NH,),SO,; pH 4.0 dialysate 1,080 15.2 36.5 39,400 2.40 68.5 28 4 (NH,),SO, Fractionation 184 22.1 139 25,600 6.29 44.6 73 5 23% Methanol precipitation 41.0 4.51 231 9,470 51.2 16.5 595 6 G-75 Sephadex chromatography 75.0 0.365 62.0 4,650 170. 8.1 1980 7 SP-Sephadex 22.5 0.273 108 2,430 396. 4.2 4600 8 DEAE-cellulose 18.0 0.111 64.4 1,160 580. 2.0 6740 9 2nd SP-Sephadex 20.0 0.070 42.5 850 605. 1.5 7030

’ Micromoles p-nitrophenol released per minute per milligram of protein when measured at 29”C, pH 4.6, using 5 mM p-nitrophenyl phosphate.


TABLE II

Amino Acid Composition of the Wheat Germ Acid Phosphatase Isoenzyme

Amino acid Quantity” Residues per molecule Nearest integer Mole %

Aspartic acid 949.6 50.2 50 9.63 Threonine * 732.2 38.7 39 7.43 Serine* 178.3 41.1 41 7.89 Glutamic acid 854.3 45.1 45 8.66 Proline 567.8 30.0 30 5.76 Glycine 1319.4 69.7 70 13.38 Alanine 872.3 46.1 46 8.85 Half-cystine* + cysteic acid 583.6 30.8 31 5.92 Valine’ 491.0 25.9 26 4.98 Methionine + sulfoxide 151.6 8.0 8 1.54 Isoleucine’ 245.1 13.0 13 2.49 Leucine 359.1 19.0 19 3.64 Tyrosine* 495.9 26.2 26 5.03 Phenylalanine 291.1 15.4 15 2.95 Histidine 360.8 19.1 19 3.66 Lysine 338.2 17.9 18 3.43 Arginine 470.1 24.8 25 4.77 Tryptophan Not determined -

Total number of amino acid residues Calculated molecular weight

’ Output from Durrum 500 amino acid analyzer; average value. * Corrected for decomposition by extrapolation of 24- and 4%hr values to zero time.

521 54,950 g/m01

’ 48-hr values.

centration of 1 mg/ml or greater was found to be approximately 3 months, while at room temperature the half- life was only 3-4 days. Extended storage requires that the samples be sterilized by filtration.

Attempts were made to lyophilize or freeze aqueous WGAP samples for storage. Activity losses (approximately 30%) were sustained during the initial freezing step. When portions of highly purified (specific activity > 200 U/mg) samples of WGAP were lyophilized, the samples lost 70-80% of their activity. Lyophilized samples at the methanol precipitate stage (Step 5) were found to retain about 50% of initial activity immediately after ly- ophilization and to lose half of the remaining activity in 3-4 months when stored at -20°C. The half-life of aqueous frozen samples (Step 5) was estimated to be about 2 weeks.

Protein concentrations below 1 mg/ml led to a rapid loss of stability, with storage in glass, polyethylene, or polypropylene vessels giving essentially the same results. The nonionic detergent, Triton X-100 and crystalline bovine serum albumin were investigated as stabilizing agents. A concentrated enzyme sample (Step 6) was diluted with 0.01 M pH 5.2 acetate buffer containing 0.1 M NaCl and either Triton X-100 or bovine serum albumin was added. The diluted samples were assayed immediately and at 24-hr intervals. The results indicate that added 1 mg/ml bovine serum albumin and 0.1% Triton X-100 are comparable in their ability to stabilize the enzyme. These

studies suggested the usefulness of adding 0.1% Triton X-100 to the chromatography buffers used in purification and the buffers used for enzyme storage. Highly purified samples (specific activity > 200 U/mg) that had a protein concentration greater than 0.5 mg/ml and which contained 0.1% Triton X-100 at pH 5.2 were generally found to have half-lives of about 6 months when stored at 4°C.

At pH 5.0 the half-life of WGAP at 4°C was 2-6 months. At pH 8.0 the enzyme was very unstable; the half-life at 25°C was approximately 40 min, which made chromatography, kinetic studies, and chemical modification experiments in this pH range difficult and indicated the need for stabilizing conditions. The possible effect of sulfhydryl group protecting reagents on stability was studied at two pH values. The effects of added 2-mercaptoethanol upon storage at 4°C at pH 5.0 and at pH 8.0 were small. At the end of a 4-week period, samples incubated with 10 mM 2-mercaptoethanol contained about 20% less activity than the control sample, while samples containing added 4.5 mM dithiothreitol lost activity at the same rate as the control sample. We conclude from these studies that neither the storage instability nor the high pH instability of WGAP is primarily due to an ox- idation of sulfhydryl groups.

The effect of prolonged dialysis against Mgt2 and EDTA at 4°C was investigated. When WGAP was dialyzed for 5 days against separate 2.0-liter portions of 0.1 M pH 5.0 acetate buffer containing either 10 mM EDTA,


9

8

L

7

2r 6

5

/ 4

-10-8-6-4-2 0 2 4 6 8 10 I/CSl

FIG. 2. Lineweaver-Burk plot illustrating substrate inhibition of WGAP-catalyzed hydrolysis of p-nitrophenyl phosphate at pH 5.0 (0); K,,, = 0.09 ? .Ol mM, V,,, = 823 f 30 mmol mg-’ min-’ and pH 6.0 (0); K,,, = 0.20 f 0.01 mM, V,,, = 956 f 12 mmol min-’ mg-‘.

10 mM Mgf2, or no addition, there was no change in the specific activity. The inability of EDTA to inactivate or of Mg+” to activate is in contrast to the acid phosphatase of wheat leaf (30). Only a slight (10 to 20%) stimulation of WGAP activity occurred in the presence of millimolar concentrations of Mg+” and Co+‘.

Molybdate at a concentration of 0.05 mM caused complete apparent inactivation. The heavy metal ions Hgt2 and Ag+ caused complete inactivation. Almost total (82%) inactivation was caused by Cu+‘, while Mnf2, Fe+2, Caf2, Fef3, and Zn+’ had little or no effect.

Additional experiments with added chelating agents also indicate that loosely bound endogeneous metal ions are not required for activity. Incubation at 25 or 4°C in the presence of EDTA and/or o-phenanthroline did not cause significantly more rapid inactivation than that caused in the absence of the chelating agents. In contrast, similar conditions could be used to inactivate sweet potato acid phosphatase with o-phenanthroline and subsequently to reactivate it with metal ions (31). We were unable to show any reactivation of the inactivated enzyme in this experiment either by added Mg+2, Mn+’ or by dialysis against Co+2. The slight stimulatory effect of Mg+2 and CO+~ and the strong inhibitory effect of molybdate are reversed upon addition of EDTA.

Steady-state kinetic studies of WGAP revealed unex- pected complexities. The nature of the reciprocal plots observed with homogeneous WGAP change as a function of pH. Substrate inhibition is observed at low pH and gives way to Michaelis-Menten kinetics near neutrality; finally, at high pH, apparent substrate activation is observed. Upward-curving reciprocal plots characteristic of substrate inhibition were observed at pH 5.0 and 6.0 with p-nitrophenyl phosphate as substrate (Fig. 2). Similar re-

sults (not shown) were obtained at pH 4.6 with glucose 6-phosphate or pyrophosphate as substrates but not with P-glycerophosphate as a substrate. At both pH 5.0 and 6.0, substrate inhibition became clearly evident only at substrate concentrations greater than 10 X K,,, and therefore might not be detected in a typical Lineweaver-Burk plot. However, a consequence of this behavior is the po- tential for serious errors in V,,, estimations when using a single high substrate concentration in an attempt to saturate the enzyme. Although such clearly evident substrate inhibition has not been reported as a kinetic property of other acid phosphatases, the determination of substrate “specificity” using a single, fixed substrate concentration is a common practice.

At pH 7.0 and 7.5 withp-nitrophenyl phosphate as the substrate, WGAP gives linear Lineweaver-Burk plots over a wide substrate concentration range (up to 400-fold at pH 7.0). Neither substrate inhibition nor substrate activation can be detected.

In the pH range of 8.0 and above, the phenomenon of substrate activation is observable (Fig. 3), with a readily apparent, downwardly curving region at high substrate concentration. At pH 8.5 and 9.0, the nonlinearity is similarly prominent. Concomitant loss of activity toward monoester (p-nitrophenyl phosphate) and diester (bis-p- nitrophenyl phosphate) substrates as well as p-nitrophenyl phenylphosphonate during the course of extensive (80%) thermal denaturation was found at pH 4.6 and 8.5. These results are consistent with activity of a single enzyme species toward all of the substrates.

The temperature dependence of the WGAP-catalyzed hydrolysis of p-nitrophenyl phosphate was investigated to obtain information about the mechanism of the enzyme. The apparent activation energy was estimated at unit pH intervals from pH 3 to 8. V,,, was determined by varying the substrate concentration at two or three temperatures. These values of V,,, were used to estimate the apparent activation energy from the slope of the plot of log vnmx versus l/T. The data at pH 5.0 show excellent linearity from 10 to 40°C (Fig. 4). At pH 3.0 (Fig. 4) and

FIG. 3. Lineweaver-Burk plot of WGAP-catalyzed hydrolysis of p- nitrophenyl phosphate at pH 8.0; Km = 0.99 _+ 0.07 mM, V,,, = 144 & 1.2 pm01 min-’ mg ‘.


l/T X IO3

FIG. 4. Arrhenius plot of the WGAP-catalyzed hydrolysis ofp-nitrophenyl phosphate. The values of k,., were determined at pH 3 (m), pH 5 (0) and pH 8 (A) as described in the text.

at pH 4.0 (data not shown) the applicability of the Ar- rhenius equation is also indicated by the linearity of the Arrhenius plot in the temperature range 15 to 37.5”C. The activation energies are given in Table III. These results indicate that at pH 5 and below the apparent activation energy is approximately constant. The linearity of the Arrhenius plot is consistent either with single rate process, or with a combination of rate constants that is independent of pH.

However, above pH 5, the apparent activation energy decreases. In Fig. 4 the three Michaelis V,,, values at pH 8.0 indicate a nonlinear dependence of log V,,, on l/T. The range of the apparent activation energies at pH 8.0 in Table III are therefore the values estimated with each extreme pair of temperatures. The values of the apparent activation energies at pH 6.0 and 7.0 are intermediate values representing estimates from data at two temperatures. A marked decrease in the apparent activation energy as the pH is increased is clearly indicated by these data. The nonlinear Arrhenius plots at high pH indicate that more than one temperature-dependent rate constant contributes to the V,,, expression (32).

Phospho group transfer is catalyzed by various acid phosphatases (2). Using highly purified WGAP, transphosphorylation was investigated because it provides a way to obtain important information about the existence and properties of a postulated covalent phosphoenzyme intermediate (33, 34).

The results in Table IV show that at high concentrations, methanol can act as a phospho group acceptor in reactions catalyzed by WGAP. Additional experiments with pH 8.5 barbital buffer in 50% methanol using 2.5, 5.0, and 10 mMp-nitrophenyl phosphate also showed that the percentage of phospho transfer varied in the range of lo-20%. Both enzymatic activity and phospho transfer were inhibited by 1.0 mM molybdate.

WGAP was inactivated by DEP over the pH range 6.0 to 8.5 (Fig. 5). To determine if inactivation led to a modification of a residue(s) at or near the active site, competitive inhibitors were added to the modification reaction mixture. The competitive inhibitors molybdate and inorganic phosphate were found to protect WGAP to some extent from inactivation by DEP. Molybdate at about 500 X Ki decreased the rate of inactivation about threefold while inorganic phosphate (50 X Ki) decreased the rate by fivefold at pH 7.0 (Fig. 5). These results suggest that the inactivation reaction involves residues at or near the active site. The initial pseudo-first-order rate constant was found to be independent of enzyme concentration (0.11 to 5.5 PM) but was linearly dependent on the DEP concentration at pH 7.4 (data not shown). This result is consistent with a bimolecular reaction between WGAP and DEP as indicated by the initial rate of inactivation, before DEP depletion.

WGAP that was substantially inactivated by treatment with DEP could be partially reactivated by 0.1 M hydroxylamine. At pH 7.4 75% inactivation was followed by 61% reactivation (Fig. 6A) while in a similar experiment at pH 7.6,91% inactivation was followed by only 39% reactivation. In an experiment at pH 8.5, 90% inactivation by DEP was rapidly achieved (2.5 min) and 65% reactivation was obtained (Fig. 6B).

The pH dependence of the rate of modification is consistent with the modification of histidyl residues. Assum- ing that DEP reacts only with the unprotonated form of histidine, then the measured rate of inactivation should depend on the state of protonation of the presumed histidine, which should vary with pH. The measured rate constant for inactivation, Kapp, should be related to the second-order rate constant, k2, for modification of unprotonated histidine, the dissociation constant of the histidine, K,, and the hydrogen ion concentration by the equation l/kapp = (l/k2 + [H+]/k,KJ. The intercept and

TABLE III

Apparent Activation Energy of the Hydrolysis of p-Nitrophenyl Phosphate by Wheat Germ Acid Phosphatase

PH Activation energy”

(kcal/mol)

3.00 12.6 4.01 12.5 5.00 12.3 6.02 10.7* 7.00 8.3’ 8.00 1.3-4.3’

a Determined from Michaelis V,,,, values. *Estimated from data at 25.0 and 37.5”C. ’ Estimates from nonlinear Arrhenius plots; the numerical values are

included for comparative purposes only.


TABLE IV

Transphosphorylation Catalyzed by Wheat Germ Acid Phosphatase”

p-Nitrophenol liberated Buffer (pH) pH of mixture * ( wol)

Acetate (3.5) 5.0 0.384 Acetate (5.0) 6.2 0.288 3,3-Dimethylglutarate (7.0) 8.6 0.388 Barbital (8.5) 9.4 0.292

D Transfer to methanol in 50% methanol. * Apparent pH as measured using a glass electrode with the 50% methanol solution.

Phosphate liberated Phospho transfer ( gmol) (%)

0.310 19.3 0.224 22.2 0.333 13.9 0.254 13.0

slope obtained from a plot of l/kapp versus [H+] (Fig. 7) was used to calculate a pK, of approximately 7.6, a value consistent with histidine.

DISCUSSION

Critical examination of previous attempts to purify WGAP suggested likely explanations of the unreprodu- cibility of earlier procedures (B-10). Comparison of the chromatographic behavior on DEAE-cellulose at pH 7.4 of a crude enzyme from Sigma Biochemical Co., an enzyme from Nutritional Biochemical Co. and the pure enzyme from Acme-Evans wheat germ as obtained here (WGAP) revealed that the relative amounts and elution positions of the phosphatase isoenzymes varies with the source. This suggested that each source may have contained the same or similar isoenzymes but in different amounts. Joyce and Grisolia (9) eluted their highly purified enzyme from pH 7.4 DEAE-cellulose at a lower ionic strength, thus distinguishing their isoenzyme from the present WGAP. Verjee (10) eluted three WGAP isoenzymes but observed no significant void volume (unre-

too

t go I- 00 z 70

z 60

Li w 50 0

5 40

TIME, min

FIG. 6. Inactivation of WGAP by diethyl pyrocarbonate and protec- tion of the enzyme by competitive inhibitors. The rate of inactivation of WGAP by diethyl pyrocarbonate was determined as described in the text: no inhibitor added (0); 50 mM inorganic phosphate added (0); and 1.0 mM molybdate added (A).

tained) peak. Under identical initial conditions, about 95% of the WGAP activity from Acme-Evans soft red wheat germ was not bound to DEAE-cellulose, indicating more anionic isoenzymes.

In order to develop a new method, samples were sur- veyed as possible starting sources. From those samples, wheat germ from Indiana grown soft red wheat supplied by Acme-Evans (Indianapolis, Indiana) was chosen because of the presence of a dominant activity band in the sample and the increased control of the source that a local supplier afforded. The strategy in the isolation of WGAP was to select a wheat germ that contained large amounts of one or two isoenzymes. The inclusion of significant details of the purification in this paper should be helpful to future workers attempting to use wheat germ as an enzyme source. The purification procedure which evolved has been thoroughly tested, but it is still more complicated than the one used to purify sunflower seed acid phosphatase to homogeneity (15). Intensive breeding of wheat and the frequent introduction of new varieties may affect the details of the purification procedure.

The purification factor in Table I is 7030. A purification factor of 11,000 would have been computed if the low

6ot \ NH20H/

- OLdI 2 45 152535455565 I 2 IO 2030405060

TIME, min TIME , min

FIG. 6. Hydroxylamine reactivation of WGAP that had been inactivated by treatment with diethyl pyrocarbonate. At the arrow, the solutions were made 0.1 M in hydroxylamine.


[HI x IO’

FIG. 7. pH dependence of the inactivation of WGAP by diethy pyrocarbonate.

specific activity water extract had been used as a basis, as was the case for the earlier procedures (8-10). Reported purification factors ranged from 15.4-fold for the most highly purified of three “isoenzymes” (10) or an estimated lo-fold purification (8), to a 3500-fold purification (9).

Comparison of specific activities is more complicated due to the different assay conditions. However, approxi- mate conversions show that two of the preparations had specific activities of less than 1 U/mg (8, 10). Using the present experimental data on the pH dependence of WGAP turnover in conjunction with the relative substrate turnover of Joyce and Grisolia (9) enables a comparison of the reported turnover of their enzyme with respect to that of WGAP. This gives comparative specific activities of 83 U/mg for the Joyce and Grisolia (9) enzyme at 25°C and pH 5.7 as compared to 600 U/mg calculated for WGAP at its pH optimum. No evidence for protein or even activity homogeneity was presented by Joyce and Grisolia (9). The specific activity of their enzyme can however be compared to that of 40 U/mg for an isoenzyme isolated earlier by us from another wheat germ source (35).

The molecular weight data indicate that WGAP occurs as a monomer with a molecular weight 58,000. This is similar to the soybean enzyme (36). Interestingly, sunflower seed acid phosphatase consists of two polypeptide chains (15), each of which is similar in size to the one that is present in the wheat germ enzyme. The finding that WGAP contains only a single polypeptide chain eliminates recombination of electrophoretically dissimilar monomer units as a structural source of some of the WGAP isoenzymes (5, 8).

The molecular weights of glycoproteins as determined by gel filtration and SDS-gel electrophoresis are not reliable if the carbohydrate content is high (37,38). Of the purified acid phosphatases that have been examined for carbohydrate content, a significant number of them have been found to be glycoproteins, including acid phosphatases from yeast, fungi, plants, and animal tissues. This widespread occurrence of carbohydrate in acid phosphatases made the carbohydrate content of WGAP of fun-

damental interest. The use of two sensitive analytical methods showed that, within the limits of experimental error, no carbohydrate was present in the pure enzyme. The size and lack of carbohydrate also distinguish this enzyme from a 35-kDa, carbohydrate-containing phos- photyrosyl-protein phosphatase from wheat seedlings that was recently described by Cheng and Tao (39).

The presence of a single polypeptide chain in WGAP, and the absence of neuraminic acid or significant amounts of other carbohydrates, serves to eliminate several of the most likely origins of some of the multiple forms of wheat germ acid phosphatase. Differences in amino acid composition and sequence represent another likely source of the multiple forms. Bread wheats are hexaploidal and are believed to contain three genomes that can be distin- guished by the electrophoretic mobility of the proteins of hexaploidal wheat seeds and the seeds of their postulated genomic origin (40). The finding that different types of wheat germs exhibited different isoenzyme patterns upon electrophoresis may reflect diverse genetic composition giving rise to different WGAP phenotypes.

WGAP, although exhibiting only mild sensitivity toward other thiol group reagents, is sensitive to Hg+’ (Ref. (9) and present results). This suggested that some degree of stability might be afforded by sulfhydryl-protecting reagents. The use of 2-mercaptoethanol as a stabilizing or protecting agent has been successful with placental acid phosphatase III (41), bovine brain acidphosphatase (42), and isoenzymes of human liver and placenta acid phosphatase (43,44). These enzymes are very sensitive to Hg+’ and thiol group reagents.

The effect of added metal ions and the metalloenzyme character of acid phosphatases has been of continual interest in the literature. There have been reports that wheat germ acidphosphatase isoenzymes contained iron (8,21). Other studies have indicated that a component of a wheat leaf acid phosphatase extract was inactivated by dialysis and reactivated by Mg+’ and other divalent metal ions (30). This indicated the need for studies on the effects of metal ions and chelating agents on the activity of purified WGAP. The rapid inactivation of WGAP by EDTA that was reported (21) and that was taken as evidence that WGAP is a metalloenzyme was not observed with the homogenous enzyme. The present results also show that the conclusion in Ref. (21) which states that molybdate irreversibly inactivates WGAP was simply due to a carry- over of molybdate into the assay mixture.

Metallo-acid phosphatases have been isolated from sweet potato, soybeans, spinach, rice, and red kidney beans (31,45). A unique property of these enzymes is the characteristic violet color attributed to the presence of binuclear iron, iron-zinc, or possibly manganese centers, and which is directly correlated with activity (46). Pre- vious workers suggested that a yellow or yellow-brown color exhibited by wheat germ acid phosphatase solutions


was due to iron (8). However, during the extensive purification which was carried out here with WGAP, the brown coloration observed in crude preparations was eliminated. No violet coloration, which seems to be characteristic of the iron or manganese in the other acid phosphatases, is exhibited by homogeneous WGAP. In fact, the spectral properties of the enzyme are similar to that of human prostatic acid phosphatase, which has been demonstrated to be a nonmetalloenzyme (47).

The nonlinear nature of the Lineweaver-Burk plots at high pH (Fig. 3) was of fundamental interest. Although many origins of this behavior are conceivable, one due to enzyme heterogeneity would represent a serious challenge to any other kinetic or enzyme specificity studies. Of the kinetic origins, those involving more than one active site would be in conflict with physical studies (present work) and stoichiometric covalent labeling of WGAP isoenzymes (48) which indicated a single polypeptide chain with one active site.

The phenomenon of downwardly curving Lineweaver- Burk plots such as that seen in Fig. 3 is widely referred to as “substrate activation.” The term implies something unusual or anomalous about the high substrate range and thus seems to suggest that the almost linear low substrate data should be used in the determination of Michaelis constants as an interference-avoidance strategy similar to that employed with substrate inhibition. It has generally been an accepted strategy to determine Michaelis constants from the unaffected low substrate range when substrate inhibition is characteristic of an enzyme sys- tem (49).

Four broad categories for the origin of apparent substrate activation were considered. (i) Artifacts of the methods or conditions of assay give apparent substrate activation (50). (ii) Two enzymes are present, both of which exhibit activity toward the substrate under consideration. (iii) A single enzyme is present which contains two active sites exhibiting negative homotropic interaction (or without interaction). (iv) A single enzyme species is present which is responsible for activity toward all of the substrates but has more than one pathway for break- down of enzyme intermediate to product.

We have carefully considered the sources of errors in assays and procedures that are enumerated in Ref. (50) and which can erroneously lead to nonlinear Lineweaver- Burk plots and concluded that the phenomenon is an in- nate, reproducible property of the WGAP enzyme.

Consideration was given to the possibility of phosphodiesterase activity at pH 8.5 which utilizedp-nitrophenyl phosphate as a substrate. Alkaline phosphodiesterases which require divalent metal ions for activity have been found in plant sources and are inactived by EDTA (51), but the addition of 1 mM EDTA did not affect the substrate activation phenomenon. The addition of 1 mM Mgt2 to p-nitrophenyl phosphate substrate in the range l-10

mM also had no effect. The possibility of more than one enzyme activity (especially diesterase activity) being responsible for apparent non-Michaelis-Menten kinetics at high pH was further investigated. Concomitant loss of activity toward monoester (p-nitrophenyl phosphate) and diester (bis-p-nitrophenyl phosphate) substrates as well as p-nitrophenyl phenylphosphonate during the course of extensive (80%) thermal denaturation was found at pH 4.6 and 8.5. These results are consistent with the presence of a single enzyme species that possesses activity toward all of the substrates.

The experimental phenomenon of substrate activation has been frequently observed in phosphatases with multiple equivalent subunits, containing one active site per subunit, including enzymes that purportedly exhibit “half of the sites reactivity”-an extreme case of negative cooperativity. A similar behavior in adenosine triphos- phatase from rat liver and acid phosphodiesterase from tobacco has been taken as evidence for two substrate binding sites in these enzymes (52, 53).

Theoretical treatments seem to preclude the observation of both substrate inhibition and substrate activation in an enzyme with two independent sites (54,55). There- fore, if the low pH substrate inhibition and high pH substrate activation exhibited by WGAP are phenomena with a common kinetic origin, either case B or case C of Harper (54,55) should apply. Case B involves two initially equivalent interacting active sites while case C includes a single active site and a single modifier site with the substrate acting as the modifier. Both of these models can exhibit either substrate activation or substrate inhibition. In the case of WGAP, the presence of only one active site is consistent with our finding of a single peptide chain per molecule and with the results of stoichiometric covalent incorporation from 32P-labeled substrate (48) so that case C appears most likely. Thus, the kinetic behavior observed at high pH with WGAP does not theoretically require two active sites. Similar kinetic behavior is exhibited by trypsin (56,57) and carboxypeptidase A (58), which have single active sites. Although generally considered unusual, it has been proposed that the property may be widespread (59,60). The present results suggest the need to examine a wide substrate range with other acid phosphatases to determine if such behavior is a general property of acid phosphatases.

Activation energies of enzyme catalyzed reactions may be complicated and difficult to interpret (32,49). However, when a single rate constant is rate limiting, a linear Ar- rhenius plot is predicted, while nonlinear Arrhenius plots typically result when two or more steps in a two-intermediate scheme have rate constants of similar magnitude but differing activation energies. The curvature would thus be analogous to that seen in mechanistic studies on prostatic phosphatase where pH variation leads to changes in the rate determining step (61). The activation


energy reported for an iron-containing porcine uterine acid phosphatase was 11.5 kcal/mol at pH 4.9 (62), while 12.5 kcal/mol was reported for prostatic acid phosphatase at pH 5.6 (63) and 10.5 kcal/mol was determined for an acid phosphatase from orange juice (64). The similarity of these values to that observed for WGAP (Table III) is interesting but probably fortuitous.

Earlier studies demonstrated the transphosphorylation capability of acid phosphatases as well as the differing specificities of various alcohols as phosphate acceptors (2). The present studies establish that homogeneous WGAP can carry out such a transphosphorylation. The results are consistent with the formation of a phosphoenzyme intermediate that can break down to products by alternate pathways utilizing either water or alcohols (here, methanol) as a phospho group acceptor. This interpre- tation is also consistent with the inactivation of the enzyme by DEP, and with the pH dependence of that inactivation. These results support the conclusion that phosphoenzyme intermediate is involved and is in fact a phosphohistidine (48).

Finally, of the several possible origins of the observed WGAP chromatographic and electrophoretic “isoenzymes” that can be considered: (i) electrophoretic or ion exchange artifacts; (ii) subunit structure with electrophoretically dissimilar subunits; (iii) varied content of neuraminic acid, (iv) genetic differences reflected in amino acid composition; (v) artifactual modification (such as partial proteolysis) during isolation; and (vi) proteolytic action prior to storage in the seed (germ), the present study provides evidence that origins (i)-(iii) are not responsible. Cloning and sequencing experiments may pro- vide the best approach to examine the remaining alter- natives.

ACKNOWLEDGMENT

This research was supported by DHHS NIH Grant GM 27003 from the National Institute of General Medical Sciences.

REFERENCES

1. Schmidt, G. (1961) in The Enzymes (Boyer, P., Ed.), Vol. 5, Academic Press, New York.

2. Hollander, V. P. (1970) in The Enzymes (Boyer, P., Ed.), Vol. 4, pp. 449-498, Academic Press, New York.

3. McLellan, W., and Lampen, J. (1963) Biochim. Biophys. Acta 67, 324-326.

4. Jacobs, M. M., Nyc, J. F., and Brown, D. M. (1971) J. Biol. Chem. 246, 1419-1425.

5. Macko, V., Honold, G., and Stahmann, M. (1967) Phytochemistry 6,465-476.

6. Shibko, S., and Tappel, A. (1963) Biochim. Biophys. Acta 73, 76- 86.

7. Matile, P., Balz, J., Semadeni, E., and Jost, M. (1965) 2. Naturforsch. 20B, 693-698.

8. Brouillard, J., and Ouellet, L. (1965) Can. J. Biochem. 43, 1899- 1905.

9. 10

11.

12.

13.

14.

15.

16.

17.

18.

19.

20.

21.

22.

23.

24.

25.

26.

27.

28.

29.

30.

31.

32.

33.

34.

35.

36.

37.

38.

39.

40. 41.

42.

43.

Joyce, B., and Grisolia, S. (1960) J. Biol. Chem. 235, 2278-2281. Verjee, Z. H. (1969) Eur. J. Biochem. 9,439-444.

Fink, A. L., and Hay, G. W. (1969) Can. J. Biochem. 47, 135-142.

McLachlan, K. D., Elliott, D. E., De Marco, D. G. (1987) Au&. J. Agric. Res. 38, 1-13.

Presley, H., and Fowden, L. (1965) Phytochemistry 4, 169-176.

MacIntyre, R., and Dean, M. (1967) Nature 214, 274-275.

Park, H. S., and Van Etten, R. L. (1986) Phytochemktry 26,351- 357.

Zhang, Z.-Y., and Van Etten, R. L. (1990) Arch. Biochem. Biophys. 282,39-49.

Tsuboi, K. K., and Hudson, P. B. (1955) Arch. Biochem. Biophys. 55, 191-205.

Rehkop, D. M., and Van Etten, R. L. (1975) Hoppe-Seylers 2. Physid. Chem. 356, 1775-1782.

Nicholson, R., and Davies, M. (1974) Eur. J. Biochem. 44, 25-35.

Saini, M. S., and Van Etten, R. L. (1978) Biochim. Biophys. Acta 526, 468-478.

Cohen, W., Bier, M., and Nord, F. F. (1958) Arch. B&hem. Biophys. 76,204-213.

Dixon, M. (1953) Biochem. J. 54, 457-458.

Chanda Rajan, J., and Klein, L. (1975) Ad. Biochem. 69, 632- 636.

Weber, K., and Osborne, M. (1969) J. Biol. Chem. 244,4406-4412.

Eckhardt, A. E., Hayes, C. E., and Goldstein, I. J. (1976) Ad. Biochem. 73,192-197.

Dubois, M., Giles, K., Hamilton, J., Rebers, P., and Smith, F. (1956) Anal. Chem. 28,350-356. Krystal, G., and Graham, A. (1976) And. Biochem. 70, 336-345.

Conchie, J., Hay, A., Strachen, I., and Levy, G. (1969) Biochem. J. 115,717-723.

Hoy, T. G., Ferdinand, N., and Harrison, P. M. (1974) Znt. J. Peptide Prot. Res. 6, 121-140.

Roberts, D. W. A. (1958) J. Biol. Chem. 230, 213-218.

Uehara, K., Fujimoto, S., Taniguchi, T., and Nakai, K. (1974) J. Biochem. 75, 639-649.

LaidIer, K. J., and Peterman, B. F. (1979) in Methods in Enzymology (Purich, D. L., Ed.), Vol. 63, pp. 234-257, Academic Press, San Diego.

Saini, M. S., Buchwald, S., Van Etten, R. L., and Knowles, J. (1981) J. Biol. Chem. 256, 10,453-10,455.

Buchwald, S., Saini, M. S., Knowles, J. R., and Van Etten, R. L. (1984) J. Biol. Chem. 259, 2208-2213. Hickey, M. E., and Van Etten, R. L. (1972) Arch. Biochem. Biophys. 152,423-455. Ullah, A. H., and Gibson, D. M. (1988) Arch. Biochem. Biophys. 260.514-520. Segrest, J. P., Jackson, R. L., Andrews, E. P., and Marchesi, V. T. (1971) Biochem. Biophys. Res. Commun. 44, 390-395. Odds, F. C., and Hierholzer, J. C. (1973) J. Bacterial. 114, 257- 266. Cheng, H.-F., and Tao, M. (1989) Biochim. Biophys. Acta 998,271- 276. Johnson, B., and Hall, 0. (1966) Acta Agri. Stand. 16, 222-224. DiPietro, D., and Zengerle, J. (1967) J. Biol. Chem. 242, 3391- 3396. Chaimovich, H., and Nome, F. (1970) Arch. Biochem. Biophys. 139, 9-16. Laidler, P. M., Taga, E. M., and Van Etten, R. L. (1982) Arch. Biochem. Biophys. 216, 512-521.


44. Waheed, A., Laidler, P. M., Wo, Y.-Y. P., and Van Etten, R. L. (1988) Biochemistry 27, 4265-4273.

45. Beck, J. L., McConachie, L. A., Summors, A. C., Arnold, W. N., DeJersey, J., and Zerner, B. (1986) Biochim. Biophys. Acta 869, 61-68.

46. Vincent, J. B., and Averill, B. A. (1990) FEBS I&t. 263,265-268.

47. Taga, E. M., Moore, D. L., and Van Etten, R. L. (1983) Prostate (N.Y.,J 4,141-150.

48. Van Etten, R. L., and Hickey, M. E. (1977) Arch. Biochem. Biophys. 183,250-259.

49. Laidler, K. J., and Bunting, P. S. (1973) The Chemical Kinetics of Enzyme Action, 2nd ed., Clarendon Press, Oxford.

50. Almond, H. R., Jr., and Niemann, C. (1960) Biochim. Biophys. Acta 44,143-750.

51. Harvey, C. I,., Olson, K. C., and Wright, R. (1970) Biochemistry 9, 921-925.

52. Ebel, R., and Lardy, H. (1975) J. Bio/. Chem. 250, 191-195.

53. Shinski, H., Miwa, M., Kato, K., Noguchi, M., Matsushima, T., and Sugimura, T. (1976) Biochemistry 15, 2185-2190.

54. Harper, E. T. (1971) J. Theor. Biol. 32, 405-414.

55. Harper, E. T. (1973) J. Theor. Biol. 39,91-10X

56. Trowbridge, C. G., Krehbiel, A., and Laskowski, M. (1963) Bio- chemistry 2, 843-850.

57. Nakata, H., and Ishii, S. (1972) J. Biochem. 72, 281-290.

58. Whitaker, J., Menger, F., and Bender (1966) Biochemistry 5, 386- 392.

59. Frieden, C. (1964) J. Biol. Chem. 239, 3522-3531.

60. Levitzki, A., and Koshland, D. E. (1969) Proc. Natl. Acad. Sci. USA 62,1121-1128.

61. Van Etten, R. L. (1982) Ann. N.Y. Acad. Sci. 390, 26-53.

62. Schlosnagle, P. C., Bazer, F. W., Tsibris, J. C., and Roberts, R. M. (1974) J. Biol. Chem. 249, 7574-7579.

63. Zhurkovskii, Y. G. (1968) Biokhimiya 33, 241-243.

64. Axelrod, B. (1947) J. Biol. Chem. 167, 57-72.

Isolation and Characterization of a Homogeneous lsoenzyme of ...

Documents