Molecular properties and active form of nonspecific acid phosphatase from Schizosaccharomyces pombe.

Equilibrium sedimentation experiments of the native acid phosphatase indicate a dimer-tetramer dissociating nonequilibrating system with a dimer Mr = 180,000 g/mol. The hydrolysis of nitrophenylphosphate was used to determine the sedimentation coefficient of the active species. The s20,w value for the species which degrades nitrophenylphosphate is 13.52 +/- 0.46 S in 1% sucrose and 13.72 +/- 0.11 S in 1.3 M sodium chloride, corresponding to the Svedberg value of the tetramer species. Several lines of evidence are presented which, together with previous data, indicate that the Schizosaccharomyces pombe nonspecific acid phosphatase is composed of 4 identical or nearly identical polypeptide chains: a, equilibrium sedimentation analysis of the enzyme in denaturing agents indicates the presence of homogeneous material having Mr = 90,800 g/mol; b, digestion with carboxypeptidase A releases 0.82 mol of tyrosine/monomer molecular weight. Concomitant phosphatase inactivation occurred during the splitting off of the tyrosyl terminal residue. Furthermore, a unique NH2-terminal residue (histidine) was determined.

' G. Dibenedetto, manuscript in preparation. seemed of great interest to investigate the molecular properties of the native acid phosphatase by sedimentation equilibrium analysis and to determine which form(s) the enzyme takes under the assay conditions. This last point was attempted by using the active enzyme-substrate sedimentation velocity method.

Materials
Norleucine was obtained from Mann Research and dansyl' chloride was from Calbiochem. Diisopropyl fluorophosphate-treated carboxypeptidase A was supplied by Worthington. Twice-distilled constant boiling HCI was used for analyses.

Methods
Nonspecific acid phosphatase from S. pombe was purified by the procedure reported previously (6) to apparent homogeneity, as ascertained from ultracentrifugal, electrophoretic, and chromatographic data. The protein concentrations and the catalytic activity of the enzyme preparations were determined according to the methods previously described (6). Gel filtration and sucrose gradient experiments were performed as reported (6).
End Group Analyses-The NHz-terminal residue was determined by the procedure reported by Gray (7). Following hydrolysis with 5.7 N HCI at 108°C in sealed, evacuated tubes for 16 h, the NH2-terminal dansyl derivative was separated and identified by thin layer chromatography on silica gel or polyamide-coated plates, using chloroform/ ethanol/acetic acid (38:4:3) for separation of acid, and butanol-l/ pyridine/acetic acid/H20 (30:20:6:24) for basic amino acid derivatives.
COOH-terminal analyses were performed by carboxypeptidase A digestion. Twelve mg of acid phosphatase (-0.4 pmol of protein monomer) dissolved in 0.1 M N-ethylmorpholineacetic acid buffer, pH 8.5, was incubated with 80 pg of diisopropyl fluorophosphate-treated carboxypeptidase A (protein molar ratio, 150:l) at 37"C, with 0.6 mol of norleucine added as internal standard. Immediately after addition of carboxypeptidase and at stated times of incubation, 2-ml aliquots were withdrawn and the protein was precipitated by addition of 1 ml of 50% trichloroacetic acid followed by 1 ml of acetone. After removal of the precipitate by centrifugation, the supernatant was extracted 3 times with an equal volume of ether, dried in a vacuum, and applied to the analyzer. In a parallel experiment, the rate of inactivation of acid phosphatase by carboxypeptidase A digestion was measured. Controls were incubated under the same conditions with boiled carboxypeptidase.
Ultracentrifugal Analyses-A partial specific volume of 0.6719 ml/ g was determined from the amino acid analysis according to Cohn and Edsall (8) and from carbohydrate content using the value of 0.65 ml/g reported for yeast mannan (9) as U of the polysaccharide, calculating the contribution to the partial specific volume from the percentage of each component. No correction of ,S was made for urea denaturation or mercaptoethanol reduction. Ultracentrifugation was performed on a Beckman model E analytical ultracentrifuge equipped with a photoelectric scanner. The absorption scanner was linked to a PDP-12 computer (Digital Equipment Corp.) for data collection and analysis.

Nonspecific Acid Phosphatase from S. pombe
Active enzyme sedimentation velocity studies were performed and calculated by the methods described by Holcenberg et al. (10) except that the double layering centerpiece was not required. The absorption of p-nitrophenol produced from p-nitrophenylphosphate was monitored at 400 nm. The relative viscosity of 0.2 M sodium succinate, pH   6.2, was determined to be 1.0364 a t 20°C. The density at the same temperature was 1.0038 ml/g. The viscosity and density of succinatebuffered sucrose and sodium chloride solutions were calculated from the International Critical Tables. Viscosity and density increments for these solutions were assumed to be additive (10). For active enzyme studies, scans were collected at 2-min intervals after reaching the experimental speed (7 to 9 min). Successive 2-min scans were used to calculate the sedimentation coefficients according to the theory of Cohen and Hahn (11) and Cohen et al. (12) as described by Holcenberg et al. (IO). The active enzyme sedimentation velocity experiments were performed under several solvent conditions and with at least two preparations of enzyme in order to determine the active form(s) of the protein. The calculations were generally made in two ways: a, the rate of movement of the maximum absorbance created by the enzyme between t +~ and to was measured as described by Holcenberg et al. (10); 6, the equivalent activity zone at 2-min differences was calculated according to the theory of Cohen and Hahn (11) as described by Holcenberg et al. (10).
Equality of these two sedimentation coefficients may be used as a criterion of homogeneity of the active species in the same way that the sedimentation coefficients of moving boundaries are used (13). Active polymers increase the weight average sedimentation coefficient while active dissociation products tend to decrease it. Based on the data of Holcenberg et al. (IO), 10% of species which are equally active as the major species, but of different oligomeric form, should be detectable. Another way of detecting active species is to change the solvent environment, perhaps inhibiting active polymers while smaller species remain active. In 1.3 M NaC1, the enzyme is only 0.6 as active as in 2% sucrose, so we used this condition to measure the active enzyme also. Finally, fractionation of the enzyme on Bio-Gel produced a trailing edge of activity, as reported in a separate paper,' which might represent small active species. A series of ultracentrifuge experiments on this "light" enzyme were conducted to determine whether small active species might exist which have been missed in dilution experiments or different solvent conditions.
All molecular weight distributions are based on high speed equilibrium experiments (13) using the six-channel cells. Rayleigh patterns were recorded on Kodak 11-G photographic plates, and the plate readings and computations were performed using computer programs developed in this laboratory (13). For the experiments on urea-dissociated enzyme, the sample was first dialyzed exhaustively against distilled water and lyophilized. Then it was dissolved in sodium acetate at pH 7.0, containing 8 M urea and 0.1 M mercaptoethanol.

RESULTS
shows the results of such an analvsis of the data from the Subunit Structure of Acid Phosphatase three cells. The average value of M z found from the number average and the weight average data is 180,000 +_ 9.000 g/mol.
mercaptoethanol: The results, obtained from the three chan-~ nels and summarized in Table I, indicate the presence of homogeneous material ( M , = -90,000) and support our previous results (6) on the homopolymeric structure of this enzyme. Amino Acid Terminal Residues-For the determination of NHz-terminal residues, 3 samples of acid phosphatase, containing 2, 4, and 10 nmol of protein monomer, were reacted with dansyl-C1 under the conditions described by Gray (7). After acid hydrolysis, the NHZ-terminal residue was separated and identified by thin layer chromatography. A single a-amino dansyl derivative was detected in each protein hydrolysate, which did not separate from standard dansyl-histidine on bidimensional chromatography.
The COOH-terminal amino acids were determined by following the rate of release with carboxypeptidase A. Only tyrosine and serine were released during the first 30 min of digestion, but the rate of release of the former amino acid was much higher than of the latter one, as expected from the specificity of this exopeptidase (14). In fact, 0.82 eq of tyrosine/ 30,000 g of protein was liberated after 30 min of incubation, as compared with 0.19 eq of serine released at the same time. This considerable difference in the rates of release of the COOH-terminal amino acids during the early incubation period allowed us to study the kinetics of inactivation of the enzyme as a function of its digestion by the exopeptidase. As shown in Fig. 1, the rate of inactivation of acid phosphatase incubated with carboxypeptidase closely matches the rate of release of the COOH-terminal tyrosine.
Molecular Properties of the Native Acid Phosphatase Sedimentation Equilibrium Studies-All the molecular weights were determined in 0.2 M sodium acetate buffer at pH 4.6, in which the protein was enzymatically active. The protein in the peak tube or in the pool of the central part of the peak (6) was analyzed in the equilibrium experiments. Essentially, the same results were obtained in three separate experiments; the results of one of such experiments are presented. Fig. 2A presents number-average molecular weight values obtained from this experiment and Fig. 2B the weight-average molecular weight. From this figure, it will be noted that molecular weight averages vary as a function of both initial concentration and the concentration point within the channel. The strong dependence of molecular weights of yeast acid phosphatase upon initial concentration shown by the lack of superimposability of molecular weight data indicates a dissociating nonequilibrating system (13). Three interpretations are possible: the first, that the system is not in chemical equilibrium; the second, that the system is in chemical equilibrium but contains a finite amount of nonequilibrating, low molecular weight material; and finally, that the system is in chemical equilibrium but the rate of equilibrium is slow in relation to the time required for the ultracentrifugal experiments. If one assumes that the data in Fin. 2  averages against a dimer molecular weight assigned: a minifigure represents the average deviations of number average molecular mum in the curve will be the molecular weight of the dimer weight data and M, data from the observed data. 0, average deviathat best fits the data experimentally obtained (13). Fig. 3 tions of M , data; 0, number average molecular weight data.

Nonspecific Acid
Phosphatase from S. pombe 3929 three cells. From these data, we conclude that the tetramer's M , = 360,000 f 18,000 g/mol.
Active Form of the Enzyme-The yeast acid phosphatase was stable throughout the course of the velocity ultracentrifugation. Table I1 presents the results of the active enzyme studies. In all cases, the volume of the layering zone was 0.01 ml. A variety of centrifuge speeds and solvent conditions are presented in the table. The light enzyme of the table was the trailing edge of fractionation on a Bio-Gel column, as reported in a separate paper.' In all cases, the buffer was 0.  ' Calculated from the peak of the' iime differences from 2-min intervals. Calculated by the method of Cohen et al. (12). succinate, pH 6.2. Sucrose or NaCl were added to the final concentration reported in the table. At pH 6.2, in 0.2 M sodium succinate buffer, the enzyme sediments with approximately the same ratio dimer/tetramer as observed at pH 4.6 in sodium acetate buffer; in this last condition, the sedimentation coefficient of the major boundary (Fig. 5, lower figure) is 14.55 f 0.4 S and the weight average sedimentation coefficient, calculated from these same experiments, is .?..o.u, = 13.35 f 0.26 S , as reported in a separate paper.' Active enzyme sedimentation velocity experiments were carried out at this pH because the product, nitrophenol, could be easily recorded by the absorption scanner at this pH. There is a slight trend of the peak sedimentation coefficients (speak) (11) to be less than the average sedimentation coefficients (saverage) (12) in sucrose, and the reverse is true in NaCl buffer. This trend is not significant in the averages of the coefficients in these solvents, nor does the light enzyme show the reverse trend in sucrose, as might be expected if small polymers of the enzyme were significantly active. We conclude from the data in this table and the patterns of time difference of the activity profiles, shown in Fig. 5 , that the tetramer is the active form of this enzyme, higher polymers may have very slight activity in sucrose, and smaller forms of the enzyme may also have slight activity, but this activity of small and large forms other than the tetramer is less than 10%. In no cases could a distinct peak of activity due to small species be detected. Occasionally, in sucrose, an indistinct zone of activity running ahead to the major activity could be found that might be due to convection or to other enzyme interactions in these runs. In any case, all experiments of Table I1 yielded enzyme activity which was constant within 10%.

DISCUSSION
The experimental data reported in this manuscript clearly confirm previous evidence indicating that nonspecific acid phosphatase from S. pombe is a polymer containing identical or very similar subunits. A single amino acid residue was found at the NH2-terminal end, the catalytic activity of the acid phosphatase was completely lost with the release of 1 tyrosine residue/protein monomer and 1 mol of Pi/protein monomer was split off. The equilibrium sedimentation experiments in 8 M urea is consistent with the presence of subunits of identical molecular weight. Since the carbohydrate accounts for 66% of the molecular weight, it is assumed that each subunit consists of a polypeptide chain weighing -30,000 that is bound to a carbohydrate weighing -60,000. Noteworthy is the loss of catalytic activity of the acid phosphatase with the release of the tyrosyl terminal residue. Some evidences that tyrosyl residues can be involved in phosphatase activities were reported. In fact, prostatic acid phosphatase was irreversibly inhibited by iodine monochloride (16). The ultraviolet spectrum of the iodinated product suggested that tyrosine was modified during the reaction and a competitive inhibitor, tartrate, protected the enzyme from inactivation. Furthermore, fructose-1,6-diphosphatase from both mammalian and yeast source is inactivated by selective modification of tyrosyl residues either by acetylation or by iodination (17). The comparative kinetics of enzyme inactivation and tyrosine splitting by carboxypeptidase A reported in Fig. 1 demonstrated that the COOH-terminal tyrosyl residue is essential for activity of acid phosphatase of S. pombe. However, further studies are required to elucidate whether it plays a direct role in catalysis or it is involved in the process of association of subunits to form the active molecular species. Since the pK value reported (18) for the phenolic hydroxyl group of tyrosine in proteins (9.8 to 10.4) is too high in comparison with the pH range of dissociation of S. pombe acid phosphatase, it seems more likely that the terminal tyrosyl residue could be directly involved in enzyme catalysis.
The equilibrium sedimentation analysis on the native enzyme at pH 4.6 revealed a dissociating, nonequilibrating system. But in spite of the lack of chemical equilibria among the molecular species, the molecular weight curves of Fig. 2 can be described as a dimer-tetramer system with a tetramer M , = 360 k 18 X 10". The 2:4 stoichiometry is confirmed by the observed "molecular space" of Fig. 4, which discriminates between monomer-dimer and dimer-tetramer models.
The active enzyme sedimentation velocity experiments reported here do not definitively prove that the tetramer of the subunit is the only active species of the protein assembly. They do establish, however, that this species is the major active form. In order to be definitive, active enzyme sedimentation velocity experiments should be carried under a variety of conditions (19), including the following: 1) a dilution series to establish the range of reliably detectable activity; 2) several solvent conditions to determine whether several active forms may exist under different solvent conditions; 3) several preparations of enzyme to establish that proteolysis during preparation is insignificant; 4) ideally, to perform the same zone sedimentation experiments on the enzyme without substrate lsphatase from S. pombe to establish that small or large species are inactive; 5 ) the amount of substrate converted to products should be constant throughout the experiment.
Of these criteria, as well as those mentioned by Holcenberg et al. (lo), we have established most. First, to find the proper conditions for these experiments, we performed a dilution range of at least 100-fold and less than 20% of the substrate was used by the enzyme zone. Second, we have employed several solvents consisting of high salt and high sucrose concentration and found the same active species under both conditions. Third, several enzyme preparations have been employed, including a fraction which would most likely exhibit activity among small species where they do exist (light enzyme of Table 11) in significant proportions. This fraction of enzyme also yielded a sedimentation coefficient consistent with the tetramer M , = 360,000 g/mol. It was not possible to perform zone sedimentation velocity experiments on the native enzyme at the same dilution as the active enzyme sedimentation runs, but the moving boundary experiments at high dilution suggest that the tetrameric species dissociates to dimers in a reversible fashion, implying that the very dilute active enzyme studies may involve a dimer -+ tetramer shift of this reversible equilibrium. Finally, it should be noted that the amount of substrate converted to p-nitrophenol in each 2-min period in the ultracentrifuge was approximately constant (within lo%), indicating that the enzyme catalyzes substrate hydrolysis and is not inhibited by the reaction product.