/ . Bwchem. 84, 193-203 (1978)
Purification and Properties of Acetate Kinase from Bacillus stearothermophilus Hiroshi NAKAJIMA, 1 Koichi SUZUKI, and Kazutomo IMAHORI Department of Biochemistry, Faculty of Medicine, The University of Tokyo, Bunkyo-ku, Tokyo 113 Received for publication, February 10, 1978
1. Acetate kinase [EC 2.7.2.1] from an thermophile, B. stearothermophilus, was purified and crystallized. 2. This enzyme was shown to be a tetramer of identical subunits which had a molecular weight of about 40,000. Amino acid analysis showed no SH group. By analyzing the CD spectrum it was deduced that this enzyme is composed of 36% ^-structure, 21 % o-helix and 43 % unordered structure. 3. This enzyme shared many common enzymatic properties with the counterpart from mesophiles, i.e. pH optimum, substrate specificity, requirement of metal ions and essential amino acid residues necessary for the catalytic activity. However, this enzyme was remarkably thermostable. 4. A plot of the reaction velocity against the concentration of acetate, ADP or acetyl phosphate gave a curve of the Michaelis-Menten type. However, such a plot against ATP gave a sigmoid curve, suggesting a homotropic allosteric nature of the enzyme. 5. From the results of chemical modification it was deduced that an amino group and an imidazole group, at least, are involved in the active site of the enzyme.
Acetate kinase (AK) [EC 2.7.2.1] is widely distributed among microorganisms (1-9). It is generally accepted that this enzyme is involved in the metabolism of pyruvate as well as in the activation of acetate. However, in anaerobic organisms this enzyme plays a significant role in energy production by catalyzing the following reaction (8, 10, 11). ADP+AcO-P — * ATP+AcO Nevertheless, this enzyme has not been sufficiently 1
Present address: Research and Development Center, YNmKA Ltd., Uji, Kyoto 611. • Abbreviations: AK, acetate kinase; AcO, acetate; AcO-P, acetyl phosphate; TNBS, trinitrobenzene sulphonic acid. Vol. 84, No. 1, 1978
193
characterized if compared with other kinases. Isolation of this enzyme from any sources has not yet been reported. The kinetic studies reported have been carried out with partially purified preparations (12-17). Physical and chemical properties have not been well-characterized except that the glutamyl residue is essential for the activity of the E. colienzym* (18) Since this failure of isolation or charactenzation of AK is ascribed to the instability of the enzyme we tried to isolate and characterized the enzyme from a thermophile, Bacillus stearotherIt is well known that almost all enzymes mophilus. obtained from a thermophile are more stable than the counterparts from a mesophile.
194
H. NAKAJIMA, K. SUZUKI, and K. IMAHORI
The present paper will report the isolation as well as molecular and enzymatic properties of AK from the thermophile.
ATP, various concentrations of ATP (0.1-20 mM) or sodium acetate (30-640 mM) and a fixed concentration of sodium acetate (400 mM) or ATP (20 HIM), respectively. The assay was carried out as above. In the standard assay system, the concenMATERIALS AND METHODS trations of sodium acetate and ATP were 400 mM Materials—B. stearothermophilus, strain NCA and 20 mM, respectively. Method (c), acetyl-P 1503, was cultured as described elsewhere (79). formed by a reverse reaction was determined by Acetate kinase (E. colt), coupling enzymes for converting it into its hydroxamate according to enzyme assays and marker proteins were obtained the method described by Rose et al. (1). The from Boehringer Mannheim, Germany. Phos- final incubation volume of 0.4 ml contained 16.7 phoenolpyruvate and nucleotides were also ob- mM ATP, 444 mM potassium acetate, 22.2 mM tained from Boehringer Mannheim, NADH and MgCl,, 389 mM hydroxylamine. The samples were NADP were products of the Oriental Yeast Co., then adjusted to the desired pH with KOH and and acetyl phosphate was the product of Sigma. the assay was initiated by addition of an approDEAE-cellulose (DE-52) was purchased from priate amount of enzyme solution to the sample Whatman, Sephadex A-50 and Ultrogel ACA-34 kept at the desired temperature. After incubation were obtained from Pharmacia and LKB, respec- for 5 min, 0.2 ml of 10% trichloroacetic acid and tively. Hydroxyapatite was prepared according 0.8 ml of 1.25% FeCl, in 1 N HCI were added. to the method of Main et al. (20). Cellulose After vigorous agitation, the amount of ferric acetate strips (Cellogel) were obtained from Cheme- chloride-hydroxamic acid complex formed was tron, Milan. measured at 540 nm. Methods—Enzyme assays: The measurements One unit of AK was defined as the amount of enzyme activity were carried out by three of the enzyme required to catalyze the formation methods to assay either the forward or reverse of 1 /*mol of ADP/min under the standard assay reaction. Method (a), synthesis of ATP from conditions. acetyl-P and ADP was determined in a coupled Determination of protein concentration: The assay with hexokinase and glucose-6-phosphate ^280*m v a ' u e f° r t n e purified thermophilic AK was dehydrogenase. The reaction mixture (0.4 ml) contained 50 mM imidazole-HCI, pH 7.3, 9.6 mM 0.90 when determined by amino acid analysis. glucose, 1.58 mM NADP, 56 units/ml hexokinase, The protein concentration was estimated spectro1.4 units/ml glucose-6-phosphate dehydrogenase, photometrically using the ^280*ra value thus ob10 mM excess of MgCl, in relation to the concen- tained. The molecular weight of the protein was tration of ADP, various concentrations of ADP assumed to be 160,000 (see " RESULTS ")• (0.1-5 mM) or acetyl-P (0.2-12 mM) and a fixed Electrophoresis: Polyacrylamide disc gel concentration of acetyl-P (12 ITIM) or ADP (5 mM), electrophoresis was carried out according to Davis respectively The assay was initiated by addition (27). SDS-polyacrylamide gelectrophoresis was of an appropriate amount of enzyme solution to the performed as described by Weber and Osborn (22). reaction mixture which was kept at 30°C by Electrophoresis of cellulose acetate strips was circulating water from a thermostat. The reaction performed in 50 mM acetate, imidazole, phosphate was followed continuously by absorbance at 340 or Tris-HCl buffer at various pH values at 20 V/cm nm in a Gilford 250 spectrophotometer. Method for 15 min at room temperature. (b), formation of ADP from sodium acetate and Modification of Amino Croups with TNBS— ATP was monitored in a coupled assay with B. stearothermophilus AK (0.066-0.32 mg/ml) was pyruvate kinase and lactate dehydrogenase. The reacted with a 11.5-144 fold molar excess of reaction mixture (0.4 ml) contained 50 mM imida- TNBS in 0.1 M borate buffer, pH 8.5, at 30°C, in zole-HCI, pH7.3, 75 mM KC1, 1.92 mM phospho- the absence or presence of the substrate, ATP, enolpyruvate, 0.24 mM NADH, 4 units/ml pyruvate ADP, AMP (final concentration: 20 mM) or acetate kinase, 11 units/ml lactate dehydrogenase, 10 mM (final concentration: 320 IITM). The remaining excess of MgQ, in relation to the concentration of activity was determined using aliquots withdrawn /. Biochem.
ACETATE KINASE FROM B. stearothermophilus
up to 50% at 0°C. After standing for 6 h, the resulting precipitate was collected by centrifugation (Step in), dissolved in 20 DIM phosphate buffer containing 2 mM EDTA, pH 6.9, and then dialyzed against the same buffer. The dialyzed sample was applied to a DEAE-cellulose column (5x43 cm) equilibrated with 20 mM phosphate buffer, pH 6.9, containing 2 mM EDTA and 50 mM KC1. The enzyme was eluted with a linear salt gradient composed of 2 liters each of the equilibration buffer and 450 mM KC1 in the same buffer. The flowrate was 20 ml/fraction/10 min. AK activity was eluted at KC1 concentrations between 0.24 and 0.29 M. Fractions containing AK activity were pooled (Step IV), and dialyzed against 70% ammonium sulfate solution, pH7.1. The resulting precipitate collected by centrifugation, was dissolved in 0.1 M phosphate buffer, pH 7.5, and dialyzed twice against 5 mM phosphate buffer, pH7.1. The dialyzed sample was applied to a column of hydroxyapatite (3.2x42 cm) equilibrated with 5 mM phosphate buffer, pH 7.1. The enzyme was eluted with a linear gradient of phosphate buffer from 5 mM to 400 mM in a total volume of 4 liters at a flow-rate of 10 ml/fraction/ 13 min. AK activity was eluted at phosphate concentrations between 0.05 and 0.06 M. The active fractions were pooled, concentrated by
L,
-
CO CNI
Vol. 84, No. 1, 1978
a
^—"" .
3 ^
l-
100 113 58.8 51.4 38.7 37.8 30.2
» Protein concentration was estimated from absorbance at 280 nm, assuming -4 280nm
ultrafiltration (Step V), and applied to a Ultrogel ACA-34 column (1.5x180 cm) equilibrated with 20 mM phosphate buffer containing 100 mM KC1, pH 7.1. The column was developed with the same buffer. The active fractions were pooled and dialyzed against 30 mM phosphate buffer containing 100 mM KCI, pH 7.5 (Step VI). The dialyzed sample was finally applied to a DEAE-Sephadex A-50 column (1.9x35 cm) equilibrated with 30 mM phosphate buffer containing 100 mM KCI, pH 7.5. The column was eluted with a linear gradient composed of 500 ml each of equilibration buffer and 500 mM KCI in the same buffer. Figure 1 shows the elution profile. The Fig. 2. Polyacrylamide gel electrophoresis of purified active fractions were pooled, concentrated by AK (50 /ig) in the absence (left) and presence (right) of ultrafiltration and stored in 50% glycerol solution. 0.1% SDS. The yield was about 30%. The specific activity of the final enzyme preparation was 1,500 position as the protein band. Furthermore, a units/mg at 30°C based on the protein concentration single symmetrical protein peak was always obcalculated using ^ao* m =0.90. We succeeded in served on ultracentrifugal analysis of this enzyme the crystallization of this enzyme from 1.28 M preparation. These results indicated that the sodium acetate solution, containing 5 mM MgC]t, preparation obtained in the present study was 50 mM imidazole-HCI, pH 7.3, and at a protein highly homogeneous. concentration of ca. 6 mg/ml. Molecular Weight and Subunit Structure—The Homogeneity—The homogeneity of the enzyme molecular weight of the enzyme preparation was preparation was examined by polyacrylamide disc determined by Ultrogel ACA-34 column chrogel electrophoresis as well as by SDS gel electro- matography using rabbit muscle pyruvate kinase, phoresis. The preparation gave a single band in yeast alcohol dehydrogenase, hog muscle lactate both cases as shown in Fig. 2. The homogeneity dehydrogenase, rabbit muscle creatine kinase and of the enzyme was also proved by electrophoresis horse heart cytochrome c as standards. In order on cellulose acetate strips at various pH values to discriminate the band of B. stearothermophilus (5.5-8.5). When the strips were stained for enzyme AK from that of E. coli AK, the assay was peractivity, a single band was detected at the same formed before and after heat treatment (60°C for / . Biochem.
ACETATE KINASE FROM B. stearothermophilus
30 min), which inactivated the latter completely. The estimated molecular weight was 160,000 ± 15,000 for the B. stearothermophilus enzyme and about 60,000 for the E. coli enzyme (3 determinations). Thus the molecular weight of the present enzyme was markedly larger than that of the E. coli enzyme (27). The molecular weight was also calculated to be 170,OO0±8,0O0 (2 determinations) from sedimentation analysis performed as described by Yphantis (28). Next the molecular weight (of subunits, if any) was determined by SDS gel electrophoresis using rabbit muscle phosphorylase, bovine serum albumin, rabbit muscle pyruvate kinase, ovalbumin, yeast alcohol dehydrogenase and hog muscle lactate dehydrogenase as standards. The estimated subunit molecular weight of the thermophilic AK was 43,000±2,000 (4 determinations). The molecular weight of E. coli AK is probably 60,0OO±3,O00 (2 determinations), if the major protein band corresponds to E. coh AK.1 Thus it is concluded that B. stearothermophilus AK is composed of 4 subunits but that E. coli AK is probably a monomeric enzyme. Amino Acid Composition—The amino acid composition results of the B. stearothermophilus enzyme are shown in Table II. The tryptophan content was estimated from the UV absorption (26). Enzyme pretreated with performic acid gave no cysteic acid. This is the first typical feature of the enzyme. Another feature is higher contents of acidic amino acid residues and lower levels of basic ones. The high content of acidic residues as well as the absence of SH groups are frequently found in other thermophilic enzymes (29, 30), though it is still uncertain whether these features are directly correlated with thermostability or not. From the amino acid composition results, the extinction coefficient of the purified enzyme, ^280nm> w a s calculated to be 0.90, assuming the molecular weight of the enzyme was 160,000. Secondary Structure of the Enzyme—The secondary structure of B. stearothermophilus AK was examined by measuring the CD spectrum in the 190-250 nm region. In the calculation of 0 * E. coli AK has not been isolated in homogeneous form, and the commercial preparation used in this study gave several bands on SDS acrylamide gel electrophoresis. Vol. 84, No. 1, 1978
197
TABLE II. Amino acid composition of B. stearothermophilus AK. Samples were hydrolyzed in vacua with 6 N HC1 for 24 h at 110°C. Amino acid
Amino acid residues (mol %)
Lys His
5.0 2.2
Number of amino acid" Residues/subunit 19 9
Arg
4.7
18
Asp
8.6
33
Thr
5 0"
19
Ser
6. 3" 10.9
24
Glu Pro
4.2
16
42
Gly
10.3
39
Ala
9.1
35
Cys/2
N.D c
Val
8.9
0 30
Met
2.7
10
He
8.2
31
Leu
7.8
30
Tyr
2.8
11
Phe
4.0
15
Trp
0.6"
2
Total
100.0
383
a Nearest integer per enzyme subunit of molecular weight 40,000. b Values corrected for destruction during acid hydrolysis, using factors of 100/94.7 (threonine) and 100/89.5 (serine), respectively (31). = Determined with a sample oxidized with performic acid. d Estimated by the method of Goodwin and Morton (26).
values, the mean residue weight (110) calculated from the amino acid composition in Table II was used. The CD spectrum is shown in Fig. 3.' The CD curve was analyzed according to the method of Greenfield and Fasman (32). The calculated 0 values at various wavelengths based on the 21 % ar-helix, 36% j9-structure, and 43% random coil conformations of poly-L-lysine fitted well with the observed CD curve in the 200-240 nm region. Substrate Specificity and Metal Requirement— The substrate specificity of the enzyme was tested by the hydroxamate assay method. As shown in Table III, ATP and GTP were more effective than UTP or CTP as phosphoryl donors. The results
H. NAKAJIMA, K. SUZUKI, and K. IMAHORI
198
TABLE IV. Metal ion specificity of B. stearothermophilus and E. coli AK's. Various metal ions were added as their chlorides to the hydroxamate assay mixture (method c) in place of Mg 1+ . Activities are expressed as % of the activity obtained with Mg*+.
0 i o
% Activity ivittai
CD I+
-10 -
250
Wavelength(nm) Fig. 3. CD profile of B. stearothermophilus AK. The spectrum was measured in 2 mM phosphate buffer, pH 7.5, in a 1 mm quartz cell at a protein concentration of 2 fiM, at 25°C. Calculated 0 values for poly-L-lysine containing 2 1 % a helix, 36% /3 structure, and 43% random coil are also indicated (O). TABLE III. Substrate specificity of B. stearothermophilus and E. coli AK's. Enzyme activities were measured by method (c). Activities are expressed as % of a control measured with ATP and acetate. % Activity to ATP-AcO-Mg system
ATP» GTPi UTP» CTP" Formate b Propionate b
B. stearo.
E. coli
100 111.5 20.3 7.3 0 5.0
100 83" 60.0 22.6 0i
Mg Mn t + Co 1+ Ca 1+ Cd' + Ni 1+ Zn«+ None
B. stearo.
E. coli
100
100 89.1 22.2 2.2 0 0.8 0 0
95.9 5.5 2.9
14.0 0.9 3.6 0
* after I h .
B. slearo. E. coli TNM HNBB TNBS 1,+KI Light oxidation EDC+GlyOMe DTNB
Cys, Tyr Cys, Trp NH, Cys, Tyr, His Cys, His, Trp, Met COOH Cys
100 100
15 0 0 50 100
100 100 20 0 0 20 30
» TNM, tetramtromethane; HNBB, 2-hydroxy-3-nitrobenzyl bromide; TNBS, 2,4,6-trinitrobenzene sulfonic acid; EDC, l-ethyl-3-(3-dimethyiaminopropyr) carbodilmide; DTNB, 5,5'-dithio bis-{2-nitrobenzoic acid); GlyOMe, glycine methyl ester.
1 2 NH2-Groups Modified (mole/suburut)
10 30 60 Reaction Time(min)
Fig. 8. Inactivation of B. stearothermophilus AK with TNBS and substrate protection. AK (2 fiM in (a), 0.41 fiM in (b)) was treated with TNBS (23 / m in (a), 59 fiM in (b)) in 0.1 M borate buffer, pH 7.5, at 30°C. (a) Correlation between the number of modified amino groups and the remaining activity of AK. The arrow shows extrapolation to zero activity, (b) Effect of substrates on the inactivation of AK. • , No substrate added; D , with 20 mM ATP; A , with 20 mM ADP; A , with 20 mM AMP; x , with 320 mM sodium acetate. Other conditions are described in " MATERIALS A N D METHODS."
J. Biochem.
ACETATE KINASE FROM B. stearothermophilus
Next, in order to rule out the possibility that the inactivation of the enzyme by TNBS is due to the concomitant conformational change of the enzyme, we compared the fluorescence spectra of the native and modified enzymes. In the latter case, excess TNBS was removed by gel filtration on Sephadex G-25. The native enzyme showed a fluorescence emission maximum at 333 nm, which did not shift appreciably upon chemical modification with TNBS. These results suggest that the conformational change of the enzyme did not occur as a result of the modification so far as judged byfluorescencespectra. Modification of Imidazole Groups with Photooxidation—The enzyme was photooxidized in the presence of Rose Bengal and the correlation between the extent of modification and the number of the modified residues was examined. After photooxidation, changes in amino acid residues were monitored by-amino acid analysis and spectrophotometric titration. The results indicated that only histidine residues decreased by photooxidation. All other residues including tryptophan, methionine and tyrosine did not decrease at all. As shown in Fig. 9 (a), disappearance of 1-1.5 histidine
201
residues per subunit resulted in the inactivation of the enzyme. The inactivation of the enzyme by photooxidation was prevented by the addition of some substrates. As shown in Fig. 9 (b) AcO or ADP afforded no protection against the inactivation, however considerable protection was provided by ATP or AcO-P. Furthermore, the protection was much more effective when these two substrates were present together. These findings suggest that the histidine residue participates in the enzyme reaction of AK. It is interesting that the presence of histidine residues in the active site of pyruvate kinase (35) was indicated by a similar experiment. Similar to the case of modification by TMBS, any large conformational change could not be detected as far as fluorescence measurements were concerned. DISCUSSION
The AK preparation obtained in the present study was highly homogeneous as judged by various tests. To the authors' knowledge this is the first example of the isolation of AK in a pure form. In the present study various molecular and catalytic properties of the purified B. stearothermophilus enzyme were examined together with the E. coli enzyme as a control. B. stearothermophilus AK resembles the E. coli enzyme and those from other sources in terms of its fundamental and catalytic properties; e.g. requirement of divalent cations, substrate specificity, Km values, pH optimum, etc. However it differs clearly from them in the following respects; i) The specific activity of the B. stearothermophilus enzyme is much higher than those from sources measured under similar 1 2 10 30 60 conditions, ii) The enzyme is far more heatHistidine Modified Reaction Time(rtna) stable, iii) B. stearothermophilus AK has a molec(mote/su burnt) Fig. 9. Photooxidation of B. stearothermophilus AK ular weight of about 160,000 and is composed of and substrate protection. AK (0.85 mg/ml in (a), 4 probably identical subunits of a molecular weight 0.03 mg/ml in (b)) was photooxidized at 30°C, in 0.1 M of 40,000. While E. coli AK is a monomeric phosphate buffer, pH 7.5 in the presence of Rose Bengal enzyme with a molecular weight of 60,000. iv) (0.0025% in (a), 0.001% in (b)). (a) Correlation be- B. stearothermophilus AK showed sigmoidal kinetics tween the number of modified histidine residues and the with respect to ATP, whereas the E. coli enzyme remaining activity of AK. (b) Effect of substrates on the inactivation of AK. • , No substrate added; O, showed Michaelis-Menten kinetics, v) Treatment with 20rnM ATP, 20 mM MgCl,, and 10 mM AcO-P, with DTNB inactivated E. coli AK, though B. A, with 10 mM AcO-P; D, with 20 HIM ATP and 20 mM stearothermophilus AK, which does not contain MgCl,; A, with 20mM ADP, x, with 320mM sodium cysteine, was not affected by the treatment. acetate. Other conditions are described in " MAIt is difficult to explain why the present enzyme TERIALS AND METHODS." is heat-stable. The present enzyme does not have Vol. 84, No. 1, 1978
202
any singular feature to explain such a property. The secondary structure of the present enzyme is comparable with those of many enzymes obtained from mesophilic sources. The tetrameric structure found in this enzyme seems to be rather unfavorable for heat stability. Only one feature exists, the absence of cysteine residues in the present enzyme. Probably the lack of free sulfhydryl groups prevents the formation of disulfide linkages between the molecules after denaturation, which is thought to be a good candidate making denaturation irreversible. Chemical modifications of the enzymes from B. stearothermophilus and E. coli show that amino, imidazole and carboxyl groups react with the respective group specific reagents with concomitant loss of activity, suggesting that these groups may be related to enzyme activity. Many reports have suggested a general essential role for these three functional groups in the function of kinase activity. For the amino group, Hollenberg et al. (34) reported that pyruvate kinsase is inactivated by TNBS, Markland et al. (37) reported that phosphoglycerate kinase is inactivated by methoxynitrotropone, Kassab et al. (38) reported the modification of creatine kinase with dansyl chloride and demonstrated that a lysine residue is essential for binding of nucleotides to the kinase. Modification of histidine residues inactivates pyruvate kinase (35), adenylate kinase (39) etc., and in these enzymes the essential histidine is thought to interact with the negatively charged phosphate moiety of the substrated. Carboxyl groups are also essential to the function of a number of kinases, i.e. phosphoglycerate kinase (36), E. coli acetate kinase (IS) etc. B. stearothermophilus and E. coli AK's are very similar to other kinases in regard to these three functional groups. Probably these three functional groups play essential roles in the catalytic function of all kinases. B. stearothermophilus AX contains no sulfhydryl group and is resistant against DTNB, while E. coli AK can be easily inactivated by this reagent. This discrepancy suggests that the sulfhydryl group does not play an essential role in the catalytic function of AX. However, such a conclusion is tentative until the E. coli enzyme is well characterized. The B. stearothermophilus enzyme shows sigmoidal kinetics with ATP. The Hill coefficient is 1.7, which suggests the presence of interaction
H. NAKAJIMA, K. SUZUKI, and K. IMAHORI
between subunits. It has been reported that the V. alcalescens (6, 40) and C. thermoaceticum (4) enzymes have sigmoidal responses to some substrates. Manuscripts describing the sigmoidal kinetic nature and more details on the subunit structure of the present enzyme are now being prepared and will be published shortly. REFERENCES 1. Rose, I.A., G-Manago, M., Korey, S.R., & Ochoa, S. (1954) J. Biol. Chem. 211, 737-756 2. Allen, S.H.G., Kellermeyer, R.W., Stjernholm, R.L., & Wood, H.G. (1964) /. Bacteriol. 87,171-187 3. Sangers, R.D., Benziman, M., & Gunsalus, I.C. (1961) J. Bacteriol. 82, 233-238 4. Schaupp, A. & Ljungdahl, L.G. (1974) Arch. Microbiol. 100, 121-129 5. Brown, M.S. & Akagi, J.M. (1966) J. Bacteriol. 92, 1273-1274 6. Pelroy, R.A. & Whiteley, H.R. (1971) /. Bacteriol. 105,259-267 7. Thorne, K.J.I. & Jones, M.E. (1963) /. Biol. Chem. 238, 2992-2998 8. Campbell, F. & Yates, M G. (1973) FEBS Lett. 37, 203-206 9. Brown, T.D.K., Pereira, C.R.S., & Stermer, F.C. (1972) /. Bacteriol. 112, 1106-1111 10. Biggins, D.R. & Dilworth, M.J. (1968) Biochim. Biophys. Acta 156, 285-296 11. Mortenson, L.E., Valentine, R.C., & Carnahan, J.E. (1963) /. Biol. Chem. 238, 794-800 12. Punch, D.L. & Fromm, H.J. (1972) Arch. Biophys. Biochem. 149, 307-315 13. Anthony, R.S. & Spector, L.B. (1971) /. Biol. Chem. 246, 6129-6135 14. Janson, C.A. & Cleland, W.W. (1974) /. Biol. Chem. 249, 2567-2571 15. Webb, R.C., Todhunter, J.A., & Punch, D.L. (1976) Arch. Biophys. Biochem. 173, 282-292 16. Skarstedt, M X & Silverstein, E. (1976) /. Biol. Chem. 251, 6775-6783 17. Anthony, R.S. & Spector, L.B. (1972) /. Biol. Chem. 247, 2120-2125 18. Todhunter, J.A. & Purich, D.L. (1974) Biochem. Biophys. Res. Commun. 60, 273-280 19. Suzuki, K. & Imahori, K. (1973) /. Biochem. 74, 955-970 20. Main, R.K., Wilkins, M.J., & Cole, LJ. (1959) /. Amer. Chem. Soc. 81," 6490-6495 21. Davis, BJ. (1964) Ann. N.Y. Acad. Sci. 121, 404-^t27 22. Weber, K. & Osborn, M. (1969) /. Biol. Chem. 244, 4406-4412 23. Goldfarb, A.R. (1966) Biochemistry 5, 2570-2578 J. Biochem.
ACETATE KINASE FROM B. stearothermophilus
203
24. Matsubara, H. & Sasaki, R.M. (1969) Biochem. 33. Boyer, P.D., (ed.) (1973) 77ie Enzymes Vol. 8, Biophys. Res. Commun. 35, 175-181 pp. 239-509, Academic Press, New York 25. Moore, S. (1963) /. Biol. Chem. 238, 235-237 34. Hollenberg, P.P., Flashner, M., & Coon, M.J. (1971) J. Biol. Chem. 246, 946-953 26. Goodwin, T.W. & Morton, R.A. (1946) Biochem. J. 40, 628-632 35. Dann, L.G. & Britton, H.G. (1974) Biochem. J. 137, 27. Anthony, R.S. & Spector, L.B. (1970) / . Biol. Chem. 405-407 245, 6739-6741 36. Brevet, A., Roustan, C , Desvages, G., Pradel, L.-A., 28. Yphantis, D.A. (1960) Ann. N.Y. Acad. Sci. 88, & Thoai, N.V. (1973) Eur. J. Biochem. 39, 141-147 586-601 37. Markland, F.S., Bacharach, A.D.E., Weber, B.H., 29. Suzuki, K. & Imahori, K. (1974) /. Biochem. 76, O'Grady, T.C., Saunders, G.C., & Umemura, N. 771-782 (1975) /. Biol. Chem. 250, 1301-1310 30. BoccO, E., Veronese, F.M., & Fontana, A. (1976) 38. Kassab, R., Roustan, C , & Pradel, L.-A. (1968) in Enzymes and Proteins from Thermophilic MicroBiochim. Biophys. Ada 167, 308-316 organisms (Zuber, H., ed.) p. 229, Birkhauser Verlag, 39. Schirmer, R.H., Schirmer, I., & Noda, L. (1970> Stuttgart Biochim. Biophys. Ada 207, 165-177 40. Bowman, CM., Valdez, R.O., & Nishimura, J.S. 31. Ress, M.W. (1946) Biochem. J. 40, 632-640 (1976) / . Biol. Chem. 251, 3117-3121 32. Greenfield, N. & Fasman, G.D. (1969) Biochemistry 8,4108-4116
Vol. 84, No. 1, 1978
Purification and Properties of Acetate Kinase from Bacillus stearothermophilus Hiroshi NAKAJIMA, 1 Koichi SUZUKI, and Kazutomo IMAHORI Department of Biochemistry, Faculty of Medicine, The University of Tokyo, Bunkyo-ku, Tokyo 113 Received for publication, February 10, 1978
1. Acetate kinase [EC 2.7.2.1] from an thermophile, B. stearothermophilus, was purified and crystallized. 2. This enzyme was shown to be a tetramer of identical subunits which had a molecular weight of about 40,000. Amino acid analysis showed no SH group. By analyzing the CD spectrum it was deduced that this enzyme is composed of 36% ^-structure, 21 % o-helix and 43 % unordered structure. 3. This enzyme shared many common enzymatic properties with the counterpart from mesophiles, i.e. pH optimum, substrate specificity, requirement of metal ions and essential amino acid residues necessary for the catalytic activity. However, this enzyme was remarkably thermostable. 4. A plot of the reaction velocity against the concentration of acetate, ADP or acetyl phosphate gave a curve of the Michaelis-Menten type. However, such a plot against ATP gave a sigmoid curve, suggesting a homotropic allosteric nature of the enzyme. 5. From the results of chemical modification it was deduced that an amino group and an imidazole group, at least, are involved in the active site of the enzyme.
Acetate kinase (AK) [EC 2.7.2.1] is widely distributed among microorganisms (1-9). It is generally accepted that this enzyme is involved in the metabolism of pyruvate as well as in the activation of acetate. However, in anaerobic organisms this enzyme plays a significant role in energy production by catalyzing the following reaction (8, 10, 11). ADP+AcO-P — * ATP+AcO Nevertheless, this enzyme has not been sufficiently 1
Present address: Research and Development Center, YNmKA Ltd., Uji, Kyoto 611. • Abbreviations: AK, acetate kinase; AcO, acetate; AcO-P, acetyl phosphate; TNBS, trinitrobenzene sulphonic acid. Vol. 84, No. 1, 1978
193
characterized if compared with other kinases. Isolation of this enzyme from any sources has not yet been reported. The kinetic studies reported have been carried out with partially purified preparations (12-17). Physical and chemical properties have not been well-characterized except that the glutamyl residue is essential for the activity of the E. colienzym* (18) Since this failure of isolation or charactenzation of AK is ascribed to the instability of the enzyme we tried to isolate and characterized the enzyme from a thermophile, Bacillus stearotherIt is well known that almost all enzymes mophilus. obtained from a thermophile are more stable than the counterparts from a mesophile.
194
H. NAKAJIMA, K. SUZUKI, and K. IMAHORI
The present paper will report the isolation as well as molecular and enzymatic properties of AK from the thermophile.
ATP, various concentrations of ATP (0.1-20 mM) or sodium acetate (30-640 mM) and a fixed concentration of sodium acetate (400 mM) or ATP (20 HIM), respectively. The assay was carried out as above. In the standard assay system, the concenMATERIALS AND METHODS trations of sodium acetate and ATP were 400 mM Materials—B. stearothermophilus, strain NCA and 20 mM, respectively. Method (c), acetyl-P 1503, was cultured as described elsewhere (79). formed by a reverse reaction was determined by Acetate kinase (E. colt), coupling enzymes for converting it into its hydroxamate according to enzyme assays and marker proteins were obtained the method described by Rose et al. (1). The from Boehringer Mannheim, Germany. Phos- final incubation volume of 0.4 ml contained 16.7 phoenolpyruvate and nucleotides were also ob- mM ATP, 444 mM potassium acetate, 22.2 mM tained from Boehringer Mannheim, NADH and MgCl,, 389 mM hydroxylamine. The samples were NADP were products of the Oriental Yeast Co., then adjusted to the desired pH with KOH and and acetyl phosphate was the product of Sigma. the assay was initiated by addition of an approDEAE-cellulose (DE-52) was purchased from priate amount of enzyme solution to the sample Whatman, Sephadex A-50 and Ultrogel ACA-34 kept at the desired temperature. After incubation were obtained from Pharmacia and LKB, respec- for 5 min, 0.2 ml of 10% trichloroacetic acid and tively. Hydroxyapatite was prepared according 0.8 ml of 1.25% FeCl, in 1 N HCI were added. to the method of Main et al. (20). Cellulose After vigorous agitation, the amount of ferric acetate strips (Cellogel) were obtained from Cheme- chloride-hydroxamic acid complex formed was tron, Milan. measured at 540 nm. Methods—Enzyme assays: The measurements One unit of AK was defined as the amount of enzyme activity were carried out by three of the enzyme required to catalyze the formation methods to assay either the forward or reverse of 1 /*mol of ADP/min under the standard assay reaction. Method (a), synthesis of ATP from conditions. acetyl-P and ADP was determined in a coupled Determination of protein concentration: The assay with hexokinase and glucose-6-phosphate ^280*m v a ' u e f° r t n e purified thermophilic AK was dehydrogenase. The reaction mixture (0.4 ml) contained 50 mM imidazole-HCI, pH 7.3, 9.6 mM 0.90 when determined by amino acid analysis. glucose, 1.58 mM NADP, 56 units/ml hexokinase, The protein concentration was estimated spectro1.4 units/ml glucose-6-phosphate dehydrogenase, photometrically using the ^280*ra value thus ob10 mM excess of MgCl, in relation to the concen- tained. The molecular weight of the protein was tration of ADP, various concentrations of ADP assumed to be 160,000 (see " RESULTS ")• (0.1-5 mM) or acetyl-P (0.2-12 mM) and a fixed Electrophoresis: Polyacrylamide disc gel concentration of acetyl-P (12 ITIM) or ADP (5 mM), electrophoresis was carried out according to Davis respectively The assay was initiated by addition (27). SDS-polyacrylamide gelectrophoresis was of an appropriate amount of enzyme solution to the performed as described by Weber and Osborn (22). reaction mixture which was kept at 30°C by Electrophoresis of cellulose acetate strips was circulating water from a thermostat. The reaction performed in 50 mM acetate, imidazole, phosphate was followed continuously by absorbance at 340 or Tris-HCl buffer at various pH values at 20 V/cm nm in a Gilford 250 spectrophotometer. Method for 15 min at room temperature. (b), formation of ADP from sodium acetate and Modification of Amino Croups with TNBS— ATP was monitored in a coupled assay with B. stearothermophilus AK (0.066-0.32 mg/ml) was pyruvate kinase and lactate dehydrogenase. The reacted with a 11.5-144 fold molar excess of reaction mixture (0.4 ml) contained 50 mM imida- TNBS in 0.1 M borate buffer, pH 8.5, at 30°C, in zole-HCI, pH7.3, 75 mM KC1, 1.92 mM phospho- the absence or presence of the substrate, ATP, enolpyruvate, 0.24 mM NADH, 4 units/ml pyruvate ADP, AMP (final concentration: 20 mM) or acetate kinase, 11 units/ml lactate dehydrogenase, 10 mM (final concentration: 320 IITM). The remaining excess of MgQ, in relation to the concentration of activity was determined using aliquots withdrawn /. Biochem.
ACETATE KINASE FROM B. stearothermophilus
up to 50% at 0°C. After standing for 6 h, the resulting precipitate was collected by centrifugation (Step in), dissolved in 20 DIM phosphate buffer containing 2 mM EDTA, pH 6.9, and then dialyzed against the same buffer. The dialyzed sample was applied to a DEAE-cellulose column (5x43 cm) equilibrated with 20 mM phosphate buffer, pH 6.9, containing 2 mM EDTA and 50 mM KC1. The enzyme was eluted with a linear salt gradient composed of 2 liters each of the equilibration buffer and 450 mM KC1 in the same buffer. The flowrate was 20 ml/fraction/10 min. AK activity was eluted at KC1 concentrations between 0.24 and 0.29 M. Fractions containing AK activity were pooled (Step IV), and dialyzed against 70% ammonium sulfate solution, pH7.1. The resulting precipitate collected by centrifugation, was dissolved in 0.1 M phosphate buffer, pH 7.5, and dialyzed twice against 5 mM phosphate buffer, pH7.1. The dialyzed sample was applied to a column of hydroxyapatite (3.2x42 cm) equilibrated with 5 mM phosphate buffer, pH 7.1. The enzyme was eluted with a linear gradient of phosphate buffer from 5 mM to 400 mM in a total volume of 4 liters at a flow-rate of 10 ml/fraction/ 13 min. AK activity was eluted at phosphate concentrations between 0.05 and 0.06 M. The active fractions were pooled, concentrated by
L,
-
CO CNI
Vol. 84, No. 1, 1978
a
^—"" .
3 ^
l-
100 113 58.8 51.4 38.7 37.8 30.2
» Protein concentration was estimated from absorbance at 280 nm, assuming -4 280nm
ultrafiltration (Step V), and applied to a Ultrogel ACA-34 column (1.5x180 cm) equilibrated with 20 mM phosphate buffer containing 100 mM KC1, pH 7.1. The column was developed with the same buffer. The active fractions were pooled and dialyzed against 30 mM phosphate buffer containing 100 mM KCI, pH 7.5 (Step VI). The dialyzed sample was finally applied to a DEAE-Sephadex A-50 column (1.9x35 cm) equilibrated with 30 mM phosphate buffer containing 100 mM KCI, pH 7.5. The column was eluted with a linear gradient composed of 500 ml each of equilibration buffer and 500 mM KCI in the same buffer. Figure 1 shows the elution profile. The Fig. 2. Polyacrylamide gel electrophoresis of purified active fractions were pooled, concentrated by AK (50 /ig) in the absence (left) and presence (right) of ultrafiltration and stored in 50% glycerol solution. 0.1% SDS. The yield was about 30%. The specific activity of the final enzyme preparation was 1,500 position as the protein band. Furthermore, a units/mg at 30°C based on the protein concentration single symmetrical protein peak was always obcalculated using ^ao* m =0.90. We succeeded in served on ultracentrifugal analysis of this enzyme the crystallization of this enzyme from 1.28 M preparation. These results indicated that the sodium acetate solution, containing 5 mM MgC]t, preparation obtained in the present study was 50 mM imidazole-HCI, pH 7.3, and at a protein highly homogeneous. concentration of ca. 6 mg/ml. Molecular Weight and Subunit Structure—The Homogeneity—The homogeneity of the enzyme molecular weight of the enzyme preparation was preparation was examined by polyacrylamide disc determined by Ultrogel ACA-34 column chrogel electrophoresis as well as by SDS gel electro- matography using rabbit muscle pyruvate kinase, phoresis. The preparation gave a single band in yeast alcohol dehydrogenase, hog muscle lactate both cases as shown in Fig. 2. The homogeneity dehydrogenase, rabbit muscle creatine kinase and of the enzyme was also proved by electrophoresis horse heart cytochrome c as standards. In order on cellulose acetate strips at various pH values to discriminate the band of B. stearothermophilus (5.5-8.5). When the strips were stained for enzyme AK from that of E. coli AK, the assay was peractivity, a single band was detected at the same formed before and after heat treatment (60°C for / . Biochem.
ACETATE KINASE FROM B. stearothermophilus
30 min), which inactivated the latter completely. The estimated molecular weight was 160,000 ± 15,000 for the B. stearothermophilus enzyme and about 60,000 for the E. coli enzyme (3 determinations). Thus the molecular weight of the present enzyme was markedly larger than that of the E. coli enzyme (27). The molecular weight was also calculated to be 170,OO0±8,0O0 (2 determinations) from sedimentation analysis performed as described by Yphantis (28). Next the molecular weight (of subunits, if any) was determined by SDS gel electrophoresis using rabbit muscle phosphorylase, bovine serum albumin, rabbit muscle pyruvate kinase, ovalbumin, yeast alcohol dehydrogenase and hog muscle lactate dehydrogenase as standards. The estimated subunit molecular weight of the thermophilic AK was 43,000±2,000 (4 determinations). The molecular weight of E. coli AK is probably 60,0OO±3,O00 (2 determinations), if the major protein band corresponds to E. coh AK.1 Thus it is concluded that B. stearothermophilus AK is composed of 4 subunits but that E. coli AK is probably a monomeric enzyme. Amino Acid Composition—The amino acid composition results of the B. stearothermophilus enzyme are shown in Table II. The tryptophan content was estimated from the UV absorption (26). Enzyme pretreated with performic acid gave no cysteic acid. This is the first typical feature of the enzyme. Another feature is higher contents of acidic amino acid residues and lower levels of basic ones. The high content of acidic residues as well as the absence of SH groups are frequently found in other thermophilic enzymes (29, 30), though it is still uncertain whether these features are directly correlated with thermostability or not. From the amino acid composition results, the extinction coefficient of the purified enzyme, ^280nm> w a s calculated to be 0.90, assuming the molecular weight of the enzyme was 160,000. Secondary Structure of the Enzyme—The secondary structure of B. stearothermophilus AK was examined by measuring the CD spectrum in the 190-250 nm region. In the calculation of 0 * E. coli AK has not been isolated in homogeneous form, and the commercial preparation used in this study gave several bands on SDS acrylamide gel electrophoresis. Vol. 84, No. 1, 1978
197
TABLE II. Amino acid composition of B. stearothermophilus AK. Samples were hydrolyzed in vacua with 6 N HC1 for 24 h at 110°C. Amino acid
Amino acid residues (mol %)
Lys His
5.0 2.2
Number of amino acid" Residues/subunit 19 9
Arg
4.7
18
Asp
8.6
33
Thr
5 0"
19
Ser
6. 3" 10.9
24
Glu Pro
4.2
16
42
Gly
10.3
39
Ala
9.1
35
Cys/2
N.D c
Val
8.9
0 30
Met
2.7
10
He
8.2
31
Leu
7.8
30
Tyr
2.8
11
Phe
4.0
15
Trp
0.6"
2
Total
100.0
383
a Nearest integer per enzyme subunit of molecular weight 40,000. b Values corrected for destruction during acid hydrolysis, using factors of 100/94.7 (threonine) and 100/89.5 (serine), respectively (31). = Determined with a sample oxidized with performic acid. d Estimated by the method of Goodwin and Morton (26).
values, the mean residue weight (110) calculated from the amino acid composition in Table II was used. The CD spectrum is shown in Fig. 3.' The CD curve was analyzed according to the method of Greenfield and Fasman (32). The calculated 0 values at various wavelengths based on the 21 % ar-helix, 36% j9-structure, and 43% random coil conformations of poly-L-lysine fitted well with the observed CD curve in the 200-240 nm region. Substrate Specificity and Metal Requirement— The substrate specificity of the enzyme was tested by the hydroxamate assay method. As shown in Table III, ATP and GTP were more effective than UTP or CTP as phosphoryl donors. The results
H. NAKAJIMA, K. SUZUKI, and K. IMAHORI
198
TABLE IV. Metal ion specificity of B. stearothermophilus and E. coli AK's. Various metal ions were added as their chlorides to the hydroxamate assay mixture (method c) in place of Mg 1+ . Activities are expressed as % of the activity obtained with Mg*+.
0 i o
% Activity ivittai
CD I+
-10 -
250
Wavelength(nm) Fig. 3. CD profile of B. stearothermophilus AK. The spectrum was measured in 2 mM phosphate buffer, pH 7.5, in a 1 mm quartz cell at a protein concentration of 2 fiM, at 25°C. Calculated 0 values for poly-L-lysine containing 2 1 % a helix, 36% /3 structure, and 43% random coil are also indicated (O). TABLE III. Substrate specificity of B. stearothermophilus and E. coli AK's. Enzyme activities were measured by method (c). Activities are expressed as % of a control measured with ATP and acetate. % Activity to ATP-AcO-Mg system
ATP» GTPi UTP» CTP" Formate b Propionate b
B. stearo.
E. coli
100 111.5 20.3 7.3 0 5.0
100 83" 60.0 22.6 0i
Mg Mn t + Co 1+ Ca 1+ Cd' + Ni 1+ Zn«+ None
B. stearo.
E. coli
100
100 89.1 22.2 2.2 0 0.8 0 0
95.9 5.5 2.9
14.0 0.9 3.6 0
* after I h .
B. slearo. E. coli TNM HNBB TNBS 1,+KI Light oxidation EDC+GlyOMe DTNB
Cys, Tyr Cys, Trp NH, Cys, Tyr, His Cys, His, Trp, Met COOH Cys
100 100
15 0 0 50 100
100 100 20 0 0 20 30
» TNM, tetramtromethane; HNBB, 2-hydroxy-3-nitrobenzyl bromide; TNBS, 2,4,6-trinitrobenzene sulfonic acid; EDC, l-ethyl-3-(3-dimethyiaminopropyr) carbodilmide; DTNB, 5,5'-dithio bis-{2-nitrobenzoic acid); GlyOMe, glycine methyl ester.
1 2 NH2-Groups Modified (mole/suburut)
10 30 60 Reaction Time(min)
Fig. 8. Inactivation of B. stearothermophilus AK with TNBS and substrate protection. AK (2 fiM in (a), 0.41 fiM in (b)) was treated with TNBS (23 / m in (a), 59 fiM in (b)) in 0.1 M borate buffer, pH 7.5, at 30°C. (a) Correlation between the number of modified amino groups and the remaining activity of AK. The arrow shows extrapolation to zero activity, (b) Effect of substrates on the inactivation of AK. • , No substrate added; D , with 20 mM ATP; A , with 20 mM ADP; A , with 20 mM AMP; x , with 320 mM sodium acetate. Other conditions are described in " MATERIALS A N D METHODS."
J. Biochem.
ACETATE KINASE FROM B. stearothermophilus
Next, in order to rule out the possibility that the inactivation of the enzyme by TNBS is due to the concomitant conformational change of the enzyme, we compared the fluorescence spectra of the native and modified enzymes. In the latter case, excess TNBS was removed by gel filtration on Sephadex G-25. The native enzyme showed a fluorescence emission maximum at 333 nm, which did not shift appreciably upon chemical modification with TNBS. These results suggest that the conformational change of the enzyme did not occur as a result of the modification so far as judged byfluorescencespectra. Modification of Imidazole Groups with Photooxidation—The enzyme was photooxidized in the presence of Rose Bengal and the correlation between the extent of modification and the number of the modified residues was examined. After photooxidation, changes in amino acid residues were monitored by-amino acid analysis and spectrophotometric titration. The results indicated that only histidine residues decreased by photooxidation. All other residues including tryptophan, methionine and tyrosine did not decrease at all. As shown in Fig. 9 (a), disappearance of 1-1.5 histidine
201
residues per subunit resulted in the inactivation of the enzyme. The inactivation of the enzyme by photooxidation was prevented by the addition of some substrates. As shown in Fig. 9 (b) AcO or ADP afforded no protection against the inactivation, however considerable protection was provided by ATP or AcO-P. Furthermore, the protection was much more effective when these two substrates were present together. These findings suggest that the histidine residue participates in the enzyme reaction of AK. It is interesting that the presence of histidine residues in the active site of pyruvate kinase (35) was indicated by a similar experiment. Similar to the case of modification by TMBS, any large conformational change could not be detected as far as fluorescence measurements were concerned. DISCUSSION
The AK preparation obtained in the present study was highly homogeneous as judged by various tests. To the authors' knowledge this is the first example of the isolation of AK in a pure form. In the present study various molecular and catalytic properties of the purified B. stearothermophilus enzyme were examined together with the E. coli enzyme as a control. B. stearothermophilus AK resembles the E. coli enzyme and those from other sources in terms of its fundamental and catalytic properties; e.g. requirement of divalent cations, substrate specificity, Km values, pH optimum, etc. However it differs clearly from them in the following respects; i) The specific activity of the B. stearothermophilus enzyme is much higher than those from sources measured under similar 1 2 10 30 60 conditions, ii) The enzyme is far more heatHistidine Modified Reaction Time(rtna) stable, iii) B. stearothermophilus AK has a molec(mote/su burnt) Fig. 9. Photooxidation of B. stearothermophilus AK ular weight of about 160,000 and is composed of and substrate protection. AK (0.85 mg/ml in (a), 4 probably identical subunits of a molecular weight 0.03 mg/ml in (b)) was photooxidized at 30°C, in 0.1 M of 40,000. While E. coli AK is a monomeric phosphate buffer, pH 7.5 in the presence of Rose Bengal enzyme with a molecular weight of 60,000. iv) (0.0025% in (a), 0.001% in (b)). (a) Correlation be- B. stearothermophilus AK showed sigmoidal kinetics tween the number of modified histidine residues and the with respect to ATP, whereas the E. coli enzyme remaining activity of AK. (b) Effect of substrates on the inactivation of AK. • , No substrate added; O, showed Michaelis-Menten kinetics, v) Treatment with 20rnM ATP, 20 mM MgCl,, and 10 mM AcO-P, with DTNB inactivated E. coli AK, though B. A, with 10 mM AcO-P; D, with 20 HIM ATP and 20 mM stearothermophilus AK, which does not contain MgCl,; A, with 20mM ADP, x, with 320mM sodium cysteine, was not affected by the treatment. acetate. Other conditions are described in " MAIt is difficult to explain why the present enzyme TERIALS AND METHODS." is heat-stable. The present enzyme does not have Vol. 84, No. 1, 1978
202
any singular feature to explain such a property. The secondary structure of the present enzyme is comparable with those of many enzymes obtained from mesophilic sources. The tetrameric structure found in this enzyme seems to be rather unfavorable for heat stability. Only one feature exists, the absence of cysteine residues in the present enzyme. Probably the lack of free sulfhydryl groups prevents the formation of disulfide linkages between the molecules after denaturation, which is thought to be a good candidate making denaturation irreversible. Chemical modifications of the enzymes from B. stearothermophilus and E. coli show that amino, imidazole and carboxyl groups react with the respective group specific reagents with concomitant loss of activity, suggesting that these groups may be related to enzyme activity. Many reports have suggested a general essential role for these three functional groups in the function of kinase activity. For the amino group, Hollenberg et al. (34) reported that pyruvate kinsase is inactivated by TNBS, Markland et al. (37) reported that phosphoglycerate kinase is inactivated by methoxynitrotropone, Kassab et al. (38) reported the modification of creatine kinase with dansyl chloride and demonstrated that a lysine residue is essential for binding of nucleotides to the kinase. Modification of histidine residues inactivates pyruvate kinase (35), adenylate kinase (39) etc., and in these enzymes the essential histidine is thought to interact with the negatively charged phosphate moiety of the substrated. Carboxyl groups are also essential to the function of a number of kinases, i.e. phosphoglycerate kinase (36), E. coli acetate kinase (IS) etc. B. stearothermophilus and E. coli AK's are very similar to other kinases in regard to these three functional groups. Probably these three functional groups play essential roles in the catalytic function of all kinases. B. stearothermophilus AX contains no sulfhydryl group and is resistant against DTNB, while E. coli AK can be easily inactivated by this reagent. This discrepancy suggests that the sulfhydryl group does not play an essential role in the catalytic function of AX. However, such a conclusion is tentative until the E. coli enzyme is well characterized. The B. stearothermophilus enzyme shows sigmoidal kinetics with ATP. The Hill coefficient is 1.7, which suggests the presence of interaction
H. NAKAJIMA, K. SUZUKI, and K. IMAHORI
between subunits. It has been reported that the V. alcalescens (6, 40) and C. thermoaceticum (4) enzymes have sigmoidal responses to some substrates. Manuscripts describing the sigmoidal kinetic nature and more details on the subunit structure of the present enzyme are now being prepared and will be published shortly. REFERENCES 1. Rose, I.A., G-Manago, M., Korey, S.R., & Ochoa, S. (1954) J. Biol. Chem. 211, 737-756 2. Allen, S.H.G., Kellermeyer, R.W., Stjernholm, R.L., & Wood, H.G. (1964) /. Bacteriol. 87,171-187 3. Sangers, R.D., Benziman, M., & Gunsalus, I.C. (1961) J. Bacteriol. 82, 233-238 4. Schaupp, A. & Ljungdahl, L.G. (1974) Arch. Microbiol. 100, 121-129 5. Brown, M.S. & Akagi, J.M. (1966) J. Bacteriol. 92, 1273-1274 6. Pelroy, R.A. & Whiteley, H.R. (1971) /. Bacteriol. 105,259-267 7. Thorne, K.J.I. & Jones, M.E. (1963) /. Biol. Chem. 238, 2992-2998 8. Campbell, F. & Yates, M G. (1973) FEBS Lett. 37, 203-206 9. Brown, T.D.K., Pereira, C.R.S., & Stermer, F.C. (1972) /. Bacteriol. 112, 1106-1111 10. Biggins, D.R. & Dilworth, M.J. (1968) Biochim. Biophys. Acta 156, 285-296 11. Mortenson, L.E., Valentine, R.C., & Carnahan, J.E. (1963) /. Biol. Chem. 238, 794-800 12. Punch, D.L. & Fromm, H.J. (1972) Arch. Biophys. Biochem. 149, 307-315 13. Anthony, R.S. & Spector, L.B. (1971) /. Biol. Chem. 246, 6129-6135 14. Janson, C.A. & Cleland, W.W. (1974) /. Biol. Chem. 249, 2567-2571 15. Webb, R.C., Todhunter, J.A., & Punch, D.L. (1976) Arch. Biophys. Biochem. 173, 282-292 16. Skarstedt, M X & Silverstein, E. (1976) /. Biol. Chem. 251, 6775-6783 17. Anthony, R.S. & Spector, L.B. (1972) /. Biol. Chem. 247, 2120-2125 18. Todhunter, J.A. & Purich, D.L. (1974) Biochem. Biophys. Res. Commun. 60, 273-280 19. Suzuki, K. & Imahori, K. (1973) /. Biochem. 74, 955-970 20. Main, R.K., Wilkins, M.J., & Cole, LJ. (1959) /. Amer. Chem. Soc. 81," 6490-6495 21. Davis, BJ. (1964) Ann. N.Y. Acad. Sci. 121, 404-^t27 22. Weber, K. & Osborn, M. (1969) /. Biol. Chem. 244, 4406-4412 23. Goldfarb, A.R. (1966) Biochemistry 5, 2570-2578 J. Biochem.
ACETATE KINASE FROM B. stearothermophilus
203
24. Matsubara, H. & Sasaki, R.M. (1969) Biochem. 33. Boyer, P.D., (ed.) (1973) 77ie Enzymes Vol. 8, Biophys. Res. Commun. 35, 175-181 pp. 239-509, Academic Press, New York 25. Moore, S. (1963) /. Biol. Chem. 238, 235-237 34. Hollenberg, P.P., Flashner, M., & Coon, M.J. (1971) J. Biol. Chem. 246, 946-953 26. Goodwin, T.W. & Morton, R.A. (1946) Biochem. J. 40, 628-632 35. Dann, L.G. & Britton, H.G. (1974) Biochem. J. 137, 27. Anthony, R.S. & Spector, L.B. (1970) / . Biol. Chem. 405-407 245, 6739-6741 36. Brevet, A., Roustan, C , Desvages, G., Pradel, L.-A., 28. Yphantis, D.A. (1960) Ann. N.Y. Acad. Sci. 88, & Thoai, N.V. (1973) Eur. J. Biochem. 39, 141-147 586-601 37. Markland, F.S., Bacharach, A.D.E., Weber, B.H., 29. Suzuki, K. & Imahori, K. (1974) /. Biochem. 76, O'Grady, T.C., Saunders, G.C., & Umemura, N. 771-782 (1975) /. Biol. Chem. 250, 1301-1310 30. BoccO, E., Veronese, F.M., & Fontana, A. (1976) 38. Kassab, R., Roustan, C , & Pradel, L.-A. (1968) in Enzymes and Proteins from Thermophilic MicroBiochim. Biophys. Ada 167, 308-316 organisms (Zuber, H., ed.) p. 229, Birkhauser Verlag, 39. Schirmer, R.H., Schirmer, I., & Noda, L. (1970> Stuttgart Biochim. Biophys. Ada 207, 165-177 40. Bowman, CM., Valdez, R.O., & Nishimura, J.S. 31. Ress, M.W. (1946) Biochem. J. 40, 632-640 (1976) / . Biol. Chem. 251, 3117-3121 32. Greenfield, N. & Fasman, G.D. (1969) Biochemistry 8,4108-4116
Vol. 84, No. 1, 1978
