Partial Purification and Characterization of Aspartate Aminotransferases from Seedling Oat Leaves

As relatively little information is available on the properties of aspartate aminotransferase from photosynthetic tissue, isolation and characterization of the two major electrophoretically distinct forms of this enzyme from seedling oat leaf homogenates were undertaken. These two forms are designated I for the more anionic form and II for the less anionic form. Form I, 80 to 90% of the total activity, has been purified to a specific activity of 120 mumol/min/mg of protein (1100-fold) and is estimated to be 90 to 95% homogeneous, as judged by analytical polyacrylamide gel electrophoresis. Form II, 10 to 20% of the total activity, has been purified to a specific activity of approximately 6 mumol/min/mg of protein (300-fold). Both forms exhibit optimal activity at pH 7.5. Michaelis constants do not differ greatly between forms I and II and are similar to those reported for the pig heart cytosolic enzyme as well as aspartate aminotransferase from other plant sources. A molecular weight of 130,000 for the purified aspartate aminotransferase I was estimated by sedimentation equilibrium centrifugation; molecular weights of the two forms are similar as estimated by sucrose density gradient centrifugation. No activation by pyridoxal phosphate has been observed during purification.


Partial
Purification

SUMMARY
As relatively little information is available on the properties of aspartate aminotransferase from photosynthetic tissue, isolation and characterization of the two major electrophoretically distinct forms of this enzyme from seedling oat leaf homogenates were undertaken. These two forms are designated I for the more anionic form and II for the less anionic form. Form I, 80 to 90% of the total activity, has been purified to a specific activity of 120 ~mol/min/mg of protein (llOO-fold) and is estimated to be 90 to 95% homogeneous, as judged by analytical polyacrylamide gel electrophoresis. Form II, 10 to 20% of the total activity, has been purified to a specific activity of approximately 6 pmol/min/ mg of protein (300-fold).
Both forms exhibit optimal activity at pH 7.5. Michaelis constants do not differ greatly between forms I and II and are similar to those reported for the pig heart cytosolic enzyme as well as aspartate aminotransferase from other plant sources. A molecular weight of 130,000 for the purified aspartate aminotransferase I was estimated by sedimentation equilibrium centrifugation; molecular weights of the two forms are similar as estimated by sucrose density gradient centrifugation. No activation by pyridoxal phosphate has been observed during purification.
Of the broad range of transamination reactions catalyzed by a variety of seedling leaf tissue extracts, including oat, the most stable and active transaminations occur between Lu-ketoglutarate and aspartate or alanine (1). Aspartate aminotransferase (EC 2.6.1.1) activity has been demonstrated to exist both inside and outside chloroplasts isolated by either aqueous or nonaqueous organelle isolation techniques (2) ; furthermore, Rehfeld and Tolbert (3) have demonstrated that this enzyme exists in at least three electrophoretically distinct forms in organelles isolated from spinach leaf homogenates. The most anionic of these is present in the highest quantity and is associated with chloroplasts and mitochondria; all three appear to be associated with peroxisomes, was immediately applied to a DEAE-cellulose column (Fig. 1) and eluted with a linear gradient of increasing ionic strength. Fractions were concentrated by pressure ultrafiltration and stored frozen in stabilizer buffer. Data are summarized in Steps 1 to 5 of Table I. Fraction 13 containing form I of the enzyme (Fig. 1) was fractionated by (NH,)&04, approximately 507, of the activity being recovered in the 50 to 55$$ (NH,)2S04 fraction. Sephades G-200 gel filtration or preparative den&y gradient ultracentrifugation accomplished a similar degree of purification but was less reproducible and more time-consuming.
Activity not appearing in the 50 to 55%) fraction was distributed among the other fractions; attempts to recover more enzyme at a higher specific activity by repeated fractionation were not successful. The (NH,),SO, fraction was well suited for final purificat.ion of form I by preparative polyacrylamide gel electrophoresis. Analytical polyacrylamide gels of the prep gel fraction containing the enzyme are shown in Fig. 2. Data for purification are summarized in Table I Further purification of form II (DEAE-cellulose Fraction A, Fig. 1) was achieved (Fig. 3) utilizing the gradient focusing technique (9). Data for purification of form II are given in Table I. Because of the low quantity of enzyme present, estimated to account for only 10 to 20% of the total activity (see Table I, footnote a), further purification has not been achieved at this time.
Characterization Studies pH Optima-Both enzyme forms exhibit optimum activity at pH 7.5 in the presence of either Tris or phosphate buffer; however, both forms are approximately 20% more active in the presence of Tris than of phosphate. Fractions between ar-TOWS were pooled to give Fraction A (aspartate aminotransferase II, specific activity of 0.34 pmol/min/mg of protein) and Fraction B (aspartate aminotransferase I, specific activity of 0.82 pmol/ min/mg of protein). Over-all recovery of activity was 74%.
Kinetic Studies-Reciprocal plots of kinetic data are shown in Figs. 4 to 6. For form II, levels of glutamate and oxalacetate were not found which would yield linear data when plotted in reciprocal form; thus, the Michaelis constants reported for this form with these substrates are estimates derived from data such as those shown in Fig. 7. No reciprocal plot for form II with oxalacetate is presented as the data were scattered and irreproducible. Due to the low Km of the enzyme for oxalacetate and low extinction coefficient for oxalacetate (1 .O MM-l cm-' in 6.1 M Tris-Cl, pH 7.5), the sensitivity of the direct assay was marginal even when 5-cm path length cuvettes and full scale measurements of 0.1 absorbance were used. At lower oxalacetate concentrations, the total absorbance of oxalacetate represented less than 0.1 A; FIG. 2. Analytical gel electrophoretic patterns for purified aspartate aminotransferase I. Gel 1, stained for activity, represents 0.5 unit of activity; Gels d and S, stained for protein, represent 10 and 50 rg of protein, respectively. thus, initial rates (10% or less of substrate utilization) could not tions; thus, these data were used to determine K, values. A be measured reliably.
summary of these results is presented in Table II. It should be noted that in all cases a reaction was observed Molecular Weight Determination-A molecular weight of only in the presence of all reactants after correction for oxalace-(128 + 9) x lo3 was calculated from two separate sedimentation tate utilization.
Rates of nonenzymatic decarboxylation of oxal-equilibrium ultracentrifugation experiments (Fig. 8). omitting acetate were determined independently for each reaction condi-the data set for the fringe closest to the rotor center. The molecution. In the direction of oxalacetate formation, rates from the lar weights of forms I and II, estimated by sucrose density gradidirect measurement of oxalacetate were very difficult to interpret ent centrifugation (Fig. 9), are identical, being 1.0 x lo5 based because of significant deviation from linearity (the rates de-on a molecular weight of 1.5 x lo5 for yeast alcohol dehydrocreasing with increasing time of reaction). The initial rates ob-genase. tained from the coupled assay where oxalacetate is continually Pyridoxal Phosphate Involvement-No activation of aspartate being removed from the reaction misture show no such devia-aminotransferase I has been observed after addition of pyridoxal phosphate to the crude supernatant, the DEAE-cellulose chromatography, or the preparative gel electrophoresis fractions.  4 (left). Initial velocity reciprocal plot for aspartate aminotransferase I with a-ketoglutarate (mM) as varied substrate, coupled assay. Velocity units are AAaao/min/I-ml reaction mixture. Enzyme concentration was 0.05 pg of protein/ml (specific activity, 120 rmol/min/mg of protein). Reaction rates were determined at 30" in l-cm path length cuvettes at these aspartate concentrations: 1, 4.00 mM; 2, 1.23 mM; S, 0.727 mM; 4, 0.516 mM; and 5, 0.400 mM. Li,~es drawn through the data points were obtained from a computer fit of the data to the ping-pong mechanism (22). min/mg of protein). Reaction rates were determined at 32" in 5-cm path length cuvettes at these oxalacetate concentrations: 1, 0.045 rnM; 2, 0.0209 mM; S, 0.0136 mM; 4, 0.0101 mM; and 6, 0.008 mM. Lines drawn through data poitzts were obtained from a computer fit of the data to the ping-pong mechanism (22). Initial velocity reciprocal plot for aspartate aminotransferase II with a-ketoglutarate (mM) as varied substrate, coupled assay. Velocity units are AAar~/min/l-ml reaction mixture. Enzyme concentration was 7.5 pg of protein/ml (specific activity, 5.0 pmol/min/mg of protein). Reaction rates were determined at 30" in l-cm path length cuvettes at these aspartate concentrations: 1, 4.00 mM; 2, 1.23 mM; 5, 0.727 mM; 4, 0.516 mrvr, and 5, 0.500 mM. Lines drawn through data points were obtained from a computer fit of the data to the ping-pong mechanism (22). FIG. 8. Sedimentation equilibrium ultracentrifugation of purified aspartate aminotransferase I in 0.10 M potassium succinate-0.01 M potassium phosphate, pH 7.5. Data were obtained from two measurements of a photograph of the interference pattern taken after 24 hours at 5.6" and 12,590 rpm. The distance from the rotor center, T, is expressed in centimeters; f is the fringe number.
ties of this enzyme from seedling oat leaf tissue have been investigated. Preliminary differential centrifugation studies indicated that oat leaves contain predominantly two electrophoretically distinct species of the enzyme. That 88% of the activity is soluble probably related to the vigorous homogenization required to disrupt plant tissue. Electrophoretic patterns indicated that only form I appears to be associated with the chloroplast-enriched fraction, whereas both forms appear in other fractions, an indication of separate compartmentation of I and II. This result agrees well with that reported by Rehfeld and Tolbert (3) for spinach. To provide information on which to base more definitive localization studies and other metabolic studies, the isolation and characterization of oat leaf aspartate aminotransferase forms were undertaken.
Purification of the two forms involves essentially two processes, separation of plant-soluble protein from low molecular weight and polyphenolic material before significant browning can occur followed by separation of the enzyme from other protein. Glycerol in the medium aids in retarding the browning reactions for a length of time sufficient to remove the protein by ammonium sulfate precipitation. Of the three gradients centrifuged simultaneously, Gradient 1 contained 37 pg of yeast alcohol dehydrogenase, Gradient 2 contained 41 Kg of aspartate aminotransferase I (specific activity, 2.8 rmol/min/mg of protein) plus 37 rg of yeast alcohol dehydrogenase, and Gradient 3 contained 38 pg of aspartate aminotransferase II (specific activity, 5.2 Fmol/min/mg of protein) plus 37 rg of yeast alcohol dehydrogenase. O-0, yeast alcohol dehydrogenase activity in Gradient 1; O-O, yeast alcohol dehydrogenase activity in Gradient 2; A--A, aspartate aminotransferase I activity in Gradient 2; m-m, aspartate aminotransferase II activity in Gradient 3 ; yeast alcohol dehydrogenase activity in Gradient 3 eluted at the same position as in Gradients 1 and 2 but is not shown.
The final preparation of form I is estimated from gel electrophoretic patterns to be 90 to 95yc homogeneous (Fig. 2); note the less anionic, Coomassie blue-positive material in the gel containing 50 pg of protein. That a low level of contaminating protein exists is supported by the nonlinearity of the sedimentation data (Fig. 8) and the presence of malate dehydrogenase activity. Although the specific activity of malate dehydrogenase seems high, 30% that of aspartate aminotransferase, the turnover number of malate dehydrogenase is probably an order of magnitude higher than for the aminotransferase, thus accounting for the low levels of contaminating protein. Form I, at a specific activity of 120 pmol/min/mg of protein, represents the most highly purified plant aspartate aminotransferase reported to date; the cauliflower enzyme has been purified to a specific activity of only 34 ~mol/min/mg of protein (4). Form II, only 10 to 20% of the total aspartate aminotransferase activity, has not yet been purified to complete homogeneity; however, a complete separation of form II from form I and a 300.fold purification has been achieved.
The primary distinguishing feature of these enzyme forms is their different electrophoretic mobilities, whereas pH optima, Michaelis constants, and molecular weights are similar for both forms. The molecular weight of 130,000 for form I is similar to that reported for aspartate aminotransferase from mammalian systems (25). Form II is not an aggregate or subunit multiple of I since both forms show identical mobilities by sucrose density gradient ultracentrifugation.
Thus, differences in mobility during polyacrylamide gel electrophoresis are not due to molecular weight differences but rather must be due to electrophoretic properties as determined by other structural features, e.g. amino acid composition.
Kinetic constants (Table II) are similar to those reported for the mammalian cytoplasmic enzyme (25) and to those reported for aminotransferases from other plant tissues (5-7). Forms I and II do not exhibit an inversion in magnitude of the K, values for aspartate and a-ketoglutarate as do the mammalian cytoplasmic and mitochondrial isoenzymes (25). The otherwise similar properties of the plant and mammalian enzymes suggest that the K, differences seen for the mammalian isozymes may not be a major consideration in the function of the malate-aspartate shuttle which could be an energy driven process (26).
Although ammonium sulfate precipitation results in substantial resolution of pyridoxal phosphate from both the pig heart cytosolic isozyme (25) and the cauliflower enzyme (4), no comparable activity loss is observed for oat leaf aspartate aminotransferase. In addition, no reactivation of the oat leaf enzyme forms by pyridoxal phosphate has been observed.