Inactivation of ATP-Dependent Deoxyribonudease of Micrococcus luteus by 2,3-Butanedione

Transcription

1 /. Biochem. 92, (1982) Inactivation of ATP-Dependent Deoxyribonudease of Micrococcus luteus by 2,3-Butanedione Itsuro NAKANO and Motoaki ANAI X Department of Medical Technology, School of Health Sciences, Kyushu University 90, Higashi-ku, Fukuoka, Fukuoka 812 Received for publication, February 24, 1982 ATP-dependent deoxyribonudease from Micrococcus luteus was purified to near homogeneity by a procedure involving gentle cell lysis, ammonium sulfate fractionation, TEAE-cellulose chromatography, Sephadex G-150 gel filtration and DNA-cellulose chromatography. Treatment of the enzyme with 2,3-butanedione, which binds specifically to arginyl residues, caused rapid loss of enzyme activities and the effect was enhanced by borate ion. The reaction obeyed first order kinetics with respect to the butanedione concentration, indicating that at least one functional arginyl residue is involved in the inactivation reaction. The enzyme was protected from inactivation by the presence of a low concentration of ATP, but not of ADP, AMP or adenosine. These results indicate that ATP-dependent deoxyribonudease of Micrococcus luteus has functional agrinyl residue(s) at an ATP-binding site. Tsuda and Strauss first reported the existence of an ATP-dependent DNase in Micrococcus luteus (7) and subsequently similar enzymes have been found in many other bacteria. This class of enzymes is thought to participate in DNA recombination and repair in several bacterial spedes (2). ATP-dependent DNase has many enzyme activities in vitro (2). To elucidate the mechanism of the diverse catalytic activities of this enzyme, we have been investigating the structures of the active centers using chemical modifiers that react with specific amino acid residues. We previously showed that there is a lysyl residue in the DNA binding site, which is independent of the ATP- binding site, of the enzymes of Bacillus laterosporus (5), Bacillus subtilis (4), and E. coli (5) using pyridoxal 5'-phosphate as a probe. In this work, we indicate that treatment of ATP-dependent DNase of M. luteus with 2,3- butanedione, which binds specifically to arginyl residues (6), affected an ATP-binding site, resulting in specific and simultaneous inactivations of ATP-dependent DNase and DNA-dependent ATPase activities. The enzyme activity was protected from this inactivation by ATP. Thus our results indicate that ATP-dependent DNase of M. luteus has functional arginyl residues in the ATP-binding site. 1 To whom reprint requests should be addressed. Abbreviations: TEAE, triethylaminocthyl (-QH4-N+- (C,H 5 ),); Hepes, A r -2-hydroxyethylpiperazine-A^'-2-ethanesulfonic acid. EXPERIMENTAL PROCEDURES Materials The strain of M. luteus and the growth conditions used were as described previ- Vol. 92, No. 4,

2 1206 I. NAKANO and M. ANAI ously (7). TEAE-celiulose was purchased from Serva (Heidelberg). Scphadex G-150 was obtained from Pharmacia. DNA-cellulose was prepared from calf thymus DNA by the method of AJberts et al. (8). «P-labeled E. coli DNA was prepared as described previously (7). DNA concentrations are expressed as nucleotide residues. [y- S! P]ATP was prepared as described previously (5). pbr322 DNA was a gift from Dr. M. Umeno of this laboratory. On agarose gel electrophoresis, the preparation contained large amounts of Form I, some Form II, and a trace amount of Form III. ATP, ADP, AMP, and adenosine were obtained from Kojin Co., Oriental Yeast Co., Kyowa Hakko Co., and Sigma, respectively. 2,3-Butanedione, glyoxal, and 1,2-cyclohexanedione were from Nakarai Chemicals, phenylglyoxal was from Tokyo Chemical Co., and Hepes was from Sigma. All other chemicals were commercial products of reagent grade. Assay of A TP-Dependent DNase This assay measures conversion of native E. coli DNA to acid-soluble nucleotides. The reaction mixture (0.3 ml) contained 66.7 ITIM Hepes buffer, ph 8.3, 30 ITIM MgClj, 8.3 ITIM 2-mercaptoethanol, 0.83 ram ATP, 20nmol of E. coli ["P]DNA and 3.5 units of enzyme. After incubation for 5 min at 37 C, acidsoluble radioactivity was determined as described previously (5). One unit of enzyme activity was defined as the amount producing 10 nmol of acidsoluble nucleotides in 30 min under these assay conditions, except that Hepes buffer was replaced by 66.7 mm glycine-naoh buffer, ph 9.4. The assay mixture for DNA-dependent ATPase (0.3 ml) contained 66.7 mm Hepes buffer, ph 8.3, 30 mm MgCl,, 8.3 mm 2-mercaptoethanol, 0.2 mm [y-»p]atp, 20 nmol of E. coli P J P]DNA and 3.5 units of enzyme. After incubation for 5 min at 37 C C, radioactivity not adsorbed to Norit was measured as described previously (7). Modification of the Enzyme The enzyme was treated with butanedione or other a-dicarbonyl reagents at 25 C. Unless otherwise noted, the preincubation mixture (0.2 ml) contained 50 mm Hepes and 50 mm borate buffers, ph 8.3, butanedione or other a-dicarbonyl reagents at the indicated concentrations, and 14 units of enzyme. After preincubation for the indicated period, the mixture was dialyzed against 100 volumes of 50 mm Hepes and 50 mm borate buffers, ph 8.3, at 0 c C for 6 h with 3 changes of the buffers to remove excess butanedione or other dicarbonyls. Then 50 fa of the dialyzate was transferred to the standard assay mixture and the remaining activity was measured. In experiments on the protective effects of adenylic acids and adenosine against butanedione inhibition, test reagents instead of butanedione were added at the indicated concentrations to the preincubation mixture. After preincubation of the mixture for 5 min, 10 mm butanedione was added and preincubation was continued for 15 min. After dialysis, remaining activities were measured as described above. Purification of ATP-Dependent DNase from M. luteus All operations were carried out at 0-4 C and all centrifugations at 23,000 x g for 20 min unless otherwise noted. The purification procedures and results of a typical purification are summarized in Table I. Cell Lysis Fifty-eight grams of frozen cells were suspended in 230 ml of Buffer A (50 mm TABLE I. Purification of ATP-dependent DNase of M. luteus. Enzyme activity was measured under standard assay conditions as described under " EXPERIMENTAL PROCEDURES." Fraction Volume (ml) Activity (units X 10-») Protein (mg) Specific activity (units/mg) Yield I. Crude extract II. Ammonium sulfate III. TEAE-celiulose IV. Sephadex G-150 V. DNA-cellulose , ,840 17, J. Biochem.

3 INACTIVATION OF OF M. luteus BY BUTANEDIONE and then applied to a DNA-cellulose column (1.0x10 cm) previously washed with Buffer D, at a flow rate of 2 ml/h. The column was washed with 50 ml of Buffer D and then a linear gradient from 0 to 0.8 M KC1 in 150 ml Buffer D was applied at a flow rate of 9 ml/h. Fractions of 3 ml were collected and active fractions, eluted with about 0.4 M KC1, were pooled (34 ml, Fraction V). Finally Fraction V was dialyzed overnight against 1 liter of Buffer E (20 mm Hepes buffer, ph 7.5, 10% glycerol, 0.1 mm EDTA, and 1 mm 2-mercaptoethanol) to remove phosphate ion which interfered with following experiments. On polyacrylamide gel electrophoresis, the final preparation gave a single major band with several faint minor bands (Fig. 1) and the position of the enzyme activity coincided with that of the major band. This purification procedure is based on the earlier procedure of Anai et al. (7). The principal modification which has allowed us to obtain the near homogeneous enzyme preparation is the replacement of sonic disruption of cells by gentle cell lysis which increases the specific activity of the extract about 20-fold, and the addition of DNA-cellulose chromatography at the end of the purification procedure. The specific activity of TEAE-Cellulose Chromatography Fraction II the final enzyme fraction was 2.5-fold higher than was applied to a TEAE-cellulose column (1.9x22 that of the previous purification (7). cm) previously equilibrated with Buffer B. After washing the column with 240 ml of 0.2 M KC1 in Buffer B, a linear gradient from 0.2 to 0.6 M KC1 in 260 ml of Buffer B was applied at a flow rate of 20 ml/h. Five-ml fractions were collected, and those with activity, elutcd with about 0.4 M KC1, were pooled (32 ml, Fraction III). Sephadex G-150 Gel Filtration Solid ammonium sulfate (11.3 g) was added to Fraction III to 50% saturation. The precipitate recovered by centrifugation was dissolved in Buffer C (Buffer B containing 0.1 M KC1) in a volume of 7.5 ml. This solution was applied to a Sephadex G-150 column (2.8 x 86 cm) previously washed with Buffer C. Fractions of 8 ml were collected at a flow rate of 30 ml/h, and those with activity, eluted immediately after the void volume, were pooled (22 ml, Fraction IV). Fig. 1. Polyacrylamide gel electrophoresis of Af. luteus DNA-Cellulose Affinity Chromatography ATP-dependent DNase. Fraction V, 20 fig, was layired a 6% gel column (0.6x7.5 cm) with 5 fi\ of 0.1% Fraction IV was dialyzed overnight against 1 liter on bromophenol blue. Electrophoresis, staining and deof Buffer D (Buffer B containing 2 mm MgCl,) staining were performed as described previously (24). Vol. 92, No. 4, 1982 Tris-HQ buffer, ph 7.5) and incubated for 30 min at 35 C with 3 ml of 10 mg/ml lysozyme. After addition of 22.5 ml of 10% streptomycin with stirring for 30 min, the suspension was centrifuged for 30 min at 18,000 x g and the supernatant was obtained (208 ml, Fraction 1). Ammonium Sulfate Fractionation Solid ammonium sulfate (59 g) was added gradually to Fraction I with stirring. After additional stirring for 60 min, the precipitate was collected by centrifugation and resuspended in 50 ml of 37 % saturated ammonium sulfate solution in Buffer A. The precipitate was collected by centrifugation. This procedure was repeated successively with 50 ml of 35 % and 25 % saturated ammonium sulfate solutions in Buffer A. Usually about 40% of the enzyme activity was recovered by extractions with 35% and 25% saturated ammonium sulfate solutions in Buffer A and the two fractions were pooled. To this combined fraction, solid ammonium sulfate was added to 50% saturation. The resulting precipitate was recovered by centrifugation, dissolved in Buffer B (20 mm potassium phosphate buffer, ph 7.5, 10% glycerol, 0.1 mm EDTA, and 1 mm 2-mercaptoethanol) in a final volume of 41.5 ml, and dialyzed overnight against 25 volumes of Buffer B (42 ml, Fraction II). 1207

4 1208 I. NAKANO and M. ANAI RESULTS Effect of 2,3-Butanedione on ATP-Dependent DNase The effects of 2,3-butanedione on the ATP-dependent DNase activity on double-stranded DNA and the double-stranded DNA-dependent ATPase activity of the enzyme were examined. The enzyme was preincubated with 10mM butanedione for the indicated periods at 25 C, and residual activities were measured and expressed as percentages of that of the sample before preincubation. Control samples were not inactivated by preincubation without butanedione. The two activities were inactivated in parallel during treatment with butanedione (Fig. 2). These results indicate butanedione inactivated the enzyme specifically and that the ATP-dependent DNase and DNA-dependent ATPase activities were coupled to each other under these conditions. Thus we used the ATP-dependent DNase activity as a representative of the enzyme activities in following experiments. Borate ion alone had no effect on the enzyme activities Preincubation time (min) Fig. 2. Effect of butanedione on ATP-dependent DNase. The enzyme (10 units) was preincubated with 10 DIM 2,3-butanedione and assayed as described under "EXPERIMENTAL PROCEDURES." Samples of 50 ft) of the dialyzate taken at the indicated times were assayed for remaining activity as described under "EXPERIMENTAL PROCEDURES" except that 0.2 mm ATP was used for DNase assay. Activities are expressed as percentages of that at zero time. ATPdependent DNase activity (#), DNA-dependent ATPase activity (O). 20 Effect of Borate Ion on Enzyme Inactivation with Butanedione The reaction of butanedione with arginyl residues in the active sites of many enzymes is known to be enhanced by the presence of borate ion (6, 9). We also confirmed this effect. When the enzyme had been preincubated with 15 mm butanedione for 20 min in 50 mm Hepes buffer, ph 8.3, 35 % of the enzyme activity remained. In contrast, only 10% of the enzyme activity remained after preincubation in 50 mm Hepes buffer, ph 8.3, containing 50 mm borate (Fig. 3). Order of Inactivation After preincubation of the enzyme with various concentrations of butanedione for certain periods, the residual activities of the enzyme were determined (Fig. 4). Plots of the logarithm of the residual activity versus the time of preincubation gave straight lines, the slopes of which depended on the concentration of butanedione. This indicates that the reaction followed pseudo-first-order kinetics with respect to time at any fixed concentration of butanedione. The order of inactivation with respect to butanedione concentration was determined by the method of Levy et al. (10) from a plot of Iogfjeciprocal of half-life] versus log[butanedione concentration]. The plot gave a straight line with a slope of 0.85 (Fig. 5). 100 Preincubation time (min) Fig. 3. Effect of borate ion on butanedione inactivation. The enzyme was preincubated with 15 mm butanedione in the presence (O) or absence ( ) of borate as described under " EXPERIMENTAL PRO- CEDURES." Borate was omitted from the dialysis buffer of samples preincubated with butanedione in the absence of borate. Activities were expressed as in Fig. 2. Control, no butanedione (A). /. Biochenu

5 INACTIVATION OF ATP-DNase OF M. luteuz BY BUTANEDIONE 1209 TABLE II. Effects of other dicarbonyl compounds. Conditions and procedures were as described under "EXPERIMENTAL PROCEDURES" and in the legend for Fig. 3. The concentration of each dicarbonyl compound tested was 10 DIM, and the preincubated time was 15 min Preincubation time (min) Fig. 4. Time-dependence of butanedione inactivation. The enzyme was treated with 0 (O), 2.5 ( ), 5 (A), 10 (A), 15 (D), or 20 ( ) IJM butanedione as described under " EXPERIMENTAL PROCEDURES." Activities were expressed as in Fig log [Butanedione] Fig. 5. Effect of butanedione concentration on enzyme inactivation. Conditions were as for Fig. 4. r 0., is the half time of inactivation. This indicates that modification of the enzyme occurred by a reaction that was first order with respect to butanedione concentration. Because butanedione is reasonably specific for arginyl residues, these results indicate that at least one arginyl residue is involved in the inactivation reaction. Effects of Other Dicarbonyl Compounds Other a-dicarbonyl compounds besides butane- Reagent None Butanedione Phenylglyoxal Glyoxal Cyclohexanedione Borate Remaining activity (%) 100 dione are known to react with arginyl residues at active sites of enzymes (11, 12). Therefore, we also modified the enzyme with phenylglyoxal, glyoxal, and 1,2-cyclohexanedione in the presence and absence of borate ion (Table II). Phenylglyoxal and butanedione inactivated the enzyme markedly, but glyoxal and cyclohexanedione had little effect on activity. When the enzyme was modified with these two reagents even in 20 mm concentration at 30 C for 15 min, no inhibitory effect was evident (data not shown). The presence of borate ion enhanced the inhibitory effect of butanedione as described above, but the inhibition by phenylglyoxal was reduced by the presence of borate ion. Protection by Adenine Nucleotides and Adenosine against Inactivation The above results indicate that ATP-dependent DNase has at least one functional arginyl residue that probably interacts with ATP or DNA, because essential arginyl residues of many enzymes are known to interact with anionic substrates or cofactors (6), and especially with ATP (12-17). Thus we examined first the protective effects of ATP, ADP, AMP, and adencsine against the inactivation by butanedione. Under conditions in which the remaining activity was 22% in the absence of protecting reagents, it was 44% in the presence of 0.1 mm ATP. No protection was observed with ADP, AMP, or Vol. 92, No. 4, 1982

6 1210 I. NAKANO and M. ANAI TABLE III. Protective effects of adenine nucleotides and adenosine against inactivation by butanedione. The conditions and procedures were as described under " EXPERIMENTAL PROCEDURES." Butanedione was added at 10 nw. Activities were expressed as percentages of that in the absence of butanedione. Addition None ATP ADP AMP Adenosine Concentration (nw) K f} & activity TABLE IV. Protective effect of DNA against inactivation by butanedione. Preincubation mixtures (100 //I, ph 8.3) contained 50 mm Hepes, 50 ITIM borate, 10 mm butanedione, 3.85 units of enzyme, and DNA at the indicated concentrations. Enzyme was preincubated with DNA for 5 min prior to addition of butanedione, and further preincubated for 15 min with butanedione at 25 C. Then a 10 n\ aliquot was transferred to the standard incubation mixture (300 fi\) and assayed for 10 min at 37 C. The final concentration of duplex linear DNA in the standard assay mixture was adjusted to 10 nmol/tube. Remaining activities were expressed as percentages of control without butanedione. Addition None E. coll DNA pbr322 DNA Concentration (mm) R! n g activity (%) adenosine at this concentration, but when added at 5 mm, ATP, ADP, AMP, and adenosine resulted in 72, 55, 45, and 33%, respectively, recovery of activity. Thus the protective effect of these reagents seemed to depend on their phosphate moiety. The results are summarized in Table III. We also examined the protective effect of DNA (Table TV). Under conditions in which the remaining activity was 18% in the absence of DNA, a slight protective effect was observed in the presence of 0.5 mm double-stranded linear DNA but no protective effect was observed in the presence of 0.01 mm DNA. A similar effect was observed with double-stranded circular DNA. Because closed or open circular duplex DNA is not a substrate of the enzyme (18), it seems unlikely that DNA specifically protects the enzyme from butanedione inactivation. This slight protective effect of DNA may be due to nonspecific stabilization of the enzyme by DNA or some interaction of guanosine in DNA, which is known to react with a-dicarbonyls (11), but the exact mechanism of the effect is unknown. These results indicate that at least one arginyl residue is located at an ATP-binding site of the enzyme. DISCUSSION To elucidate the catalytic mechanism of the diverse activities of ATP-dependent DNase of M. luteus, we have been investigating the structures of the active centers using chemical modifiers that are specific for certain amino acid residues. The results presented here indicate that functional arginyl residue(s) are present in an ATP-binding site of the enzyme. Many enzymes reacting with anionic substrates or cofactors are known to have arginyl residues at their ligand binding sites (6), and essential arginyl residues have been found to be present in the ATP-binding sites of some ATPdependent enzymes (12-17). It is well established that 2,3-butanedione reversibly modifies arginyl residues with reasonable specificity-under mildconditions and that the reaction is enhanced by the presence of borate ion (6, 9). ATP-dependent DNase of M. luteus was rapidly inactivated by 2,3-butanedione, the inactivation was augmented by borate, and the reaction /. Biochem.

7 IN ACTIVATION OF ATP-DNase OF M. luteus BY BUTANEDIONE 1211 followed first order kinetics with respect to butanedione concentration. These results indicated that at least one functional arginyl residue was involved in the inactivation reaction by butanedione. Effects of other dicarbonyls were various (Table II). Effectiveness of these reagents is known to vary with enzyme species as well as with reaction conditions. Similar results were reported by Riordan (79). Rapid inactivation of our enzyme by phenylglyoxal also offered supporting evidence that the inactivation was due to specific modification of arginyl residues of the enzyme. The effects of borate ion on inactivation were different between phenylglyoxal and butanedione (Table II). Similar results were reported by Daemen and Riordan (20), and Rogers et al. (21). Butanedione reacts with the guanido group of arginine to form cw-diol,dihydroxyimidazoline derivatives first, with which borate reacts to form a stable complex. Thus, borate is considered to stabilize the butanedione-arginine complex (19). On the other hand, phenylglyoxal is an a-ketoaldehyde, and two molecules of phenylglyoxal are known to react with one molecule of arginyl residue. The c«-diol,dihydroxyimidazoline derivatives, formed by condensation of the first phenylglyoxal and arginine, further reacts with the aldehyde of a second phenylglyoxal to form a stable complex (22). Thus, borate ion is not required for stabilization of the phenylglyoxalarginine complex. Furthermore, excess borate is rather inhibitory to the reaction because the formation of borate-phenylglyoxal complex results in reduction of the effective phenylglyoxal concentration (27). The presence of 0.1 mm ATP protected the enzyme from inactivation by 10 mm butanedione significantly. With increase in concentration to 5 mm, ADP, AMP, and even adenosine protected the enzyme, though to a lesser extent than ATP, and the protective effect appeared to depend on the amount of phosphate moiety of the protecting reagents; that is, the effectiveness of the compounds was in the order ATP > ADP > AMP > adenosine. These results strongly suggest that the guanido group of arginyl residue(s) at an active site recognizes the phosphate moiety, an anionic group of ATP, and that the functional arginyl residues are present in an ATP-binding site. The observation that 5 mm adenosine, which has no phosphate group, protected the enzyme could be explained by supposing the existence of an adenine or adenosine recognition site distinct from the phosphate recognition site: adenosine, bound to the enzyme, could interfere with the access of butanedione to the phosphate recognition site by steric hindrance, resulting in protection of the enzyme. Our previous studies (23) showed that ATP and datp were the most effective as cofactors of this enzyme activity, that (d)gtp and ITP were one-tenth as effective as (d)atp but more effective than other (d)ntps, and that the triphosphate group was essential for activity. These observations imply that an ATP-binding site of the enzyme recognizes the adenine base and triphosphate group of ATP separately, which is consistent with the present results. We thank Professor Y. Takagi for providing facilities and for his encouragement during this work, Mr. M. Hara and other members of the Laboratories of Kaken Chemical Co., Ltd. for generous supply of bacterial cells and Dr. J. Nakayama and Dr. T. Fujiyoshi for helpful discussions. REFERENCES 1. Tsuda, Y. & Strauss, B.S. (1964) Biochemistry 3, Whitehead, E.P. (1979) in Macromolecules in the Functioning Cell (Salvatore, F., Marino, G., & Volpe, P., eds.) pp , Plenum Publishing Corporation, New York 3. Fujiyoshi, T., Nakayama, J., & Anai, M. (1981) /. Biochem. 89, Eshima, N., Nakano, I., & Anai, M. (1979) Seikagaku (in Japanese) 51, Anai, M., Fujiyoshi, T., Nakayama, J., & Takagi, Y. (1979) /. Biol. Chem. 254, Riordan, J.F., McElvany, K.D., & Borders, C.L., Jr. (1977) Science 195, Anai, M., Hirahashi, T., & Takagi, Y. (1970) /. Biol. Chem. 245, Alberts, B.M., Amodio, F.J., Jenkins, M., Gutmann, E.D., & Ferris, F.L. (1968) Cold Spring Harbor Symp. Quant. Biol. 33, Borders, C.L., Jr., Riordan, J.F., & Auld, D.S. (1975) Biochem. Biophys. Res. Commun. 66, Levy, H.M., Leber, P.D., & Ryan, E.M. (1963) J. Biol. Chem. 238, Vol. 92, No. 4, 1982

8 Patthy, L. & Thesz, J. (1980) Eur. J. Biochem. 105, Franks, DJ., Tunnicliff, G., & Ngo, T.T. (1980) Biochim. Biophys. Acta 611, Powers, S.G. & Riordan, J.F. (1975) Proc. Natl. Acad. Sci. U.S. 11, BerghSuser, J. (1975) Biochim. Biophys. Acta 397, Borders, C.L., Jr. & Riordan, J.F. (1975) Biochemistry 14, Depont, J.J.H.H.M., Schoot, B.M., Van Prooijen- Van Eedcn, A., & Bonting, S.L. (1977) Biochim. Biophys. Acta 482, Morkin, E., Flink, I.L., & Banerjee, S.K. (1979) /. Biol. Chem. 254, I. NAKANO and M. ANAI 18. Takagi, Y., Matsubara, K. F & Anai, M. (1972) Biochim. Biophys. Acta 269, Riordan, J.F. (1973) Biochemistry 12, Daemen, F.J.M. & Riordan, J.F. (1974) Biochemistry 13, Rogers, T.B., Barrcsen, T., & Feeney, R.E. (1978) Biochemistry 17', Takahashi, K. (1968) /. Biol. Chem. 243, Anai, M., Hirahashi, T., Yamanaka, M., & Takagi, Y. (1970) /. Biol. Chem. 245, Fujiyoshi, T. & Anai, M. (1981) /. Biochem. 89, Biochem.