Inactivation precedes conformation change during thermal denaturation of adenylate kinase

Biochimica et Biophysica Acta, 1164(1993)61-67

61

© 1993 Elsevier Science Publishers B.V. All rights reserved 0167-4838/93/$06.00

BBAPRO 34515

Inactivation precedes conformation change during thermal denaturation of adenylate kinase Yan-Ling Zhang, Jun-Mei Zhou and Chen-Lu Tsou National Laboratory of Biomacromolecules, Institute of Biophysics, Academia Sinica, Beijing (China)

(Received 8 November 1992) (Revised manuscript received 9 February 1993)

Key words: Adenylate kinase; Thermal denaturation; Protein unfolding;Protein refolding;Inactivation; Reactivation During the thermal denaturation of rabbit muscle adenylate kinase, the extents and rates of both unfolding and aggregation are dependent on protein concentration. Under identical conditions, inactivation takes place at a lower temperature than noticeable conformational changes and aggregation as measured by fluorescence, second derivative absorption spectroscopy, far ultraviolet circular dichroism and light scattering. Kinetics of inactivation can be resolved into two phases and at the same protein concentrations, the unfolding and aggregation rates are about one order of magnitude slower than the fast phase and approximately the same as the slow phase rate of the inactivation reaction between 35 and 60°C. This is in general accord with the suggestion made previously that the active site of this enzyme is situated in a region more flexible than the molecule as a whole (Tsou, C.L. (1986) Trends Biochem. Sci. 11, 427-429). The inactivated enzyme cannot be reactivated by cooling and standing at 4°C but can be over 80% reactivated by cooling and first standing in 3 M guanidine hydrochloride followed by diluting out the denaturant.

Introduction

Adenylate kinase (EC 2.7.4.3) is an important enzyme in energy metabolism catalyzing the phosphorylation of AMP by ATP forming two A D P molecules [1]. The muscle cytosolic enzyme containing 194 amino acids in a single peptide chain is a relatively small enzyme among the kinases with a known sequence and three-dimensional structure [2]. The substrate binding and mechanism of action of this enzyme are fairly well understood [3-8]. It has two structural domains responsible, respectively, for A T P and AMP binding [9,10] which lead to conformational changes of the enzyme molecule. However, the conformational and activity changes during unfolding and refolding of this enzyme and the importance of conformational integrity on its activity have been but little explored. In this laboratory, the comparison of activity and conformational changes during denaturation of creatine kinase [12,13], glyceraldehyde-3-phosphate dehydrogenase [14,15], ribonuclease [16] and lactate dehydrogenase [17] have been examined. During unfolding of these enzymes by guanidine hydrochloride, urea and

Correspondence to: Institute of Biophysics, Academia Siniea, 15 Datun Road, Beijing 100101, China.

heat, inactivation invariably occurs before significant conformational change so far as can be detected by a number of physical and chemical techniques. It has therefore been proposed that the active sites of these enzymes are situated in molecular regions more flexible than the molecules as a whole [18]. Similar results have been reported recently from different laboratories for fumarase [19], citrate synthase [20] and other enzymes. The purpose of this paper is to compare activity and conformational changes during thermal unfolding of adenylate kinase, as well as the refolding of the thermally-denatured enzyme. The thermal denaturation of adenylate kinase from Escherichia coli has been briefly examined by Reinstein et al. [21]. The use of physical factors like heat to induce unfolding in the present series of studies has been prompted by the suggestion that the inactivation by guanidine hydrochloride or urea could be due to binding of the denaturants as inhibitors [22]. Like the thermal denaturation of glyceraldehyde-3-phosphate dehydrogenase [23], inactivation of adenylate kinase occurs before aggregation and noticeable conformational change. Inactivation, denaturation and aggregation are irreversible; reactivation and refolding are possible only after cooling and disaggregation of the thermally denatured enzyme in guanidine hydrochloride followed by further dilution with buffer.

62 Materials and Methods

Reagents. Glucose-6-phosphate dehydrogenase, hexokinase and NADP were Sigma products, guanidine hydrochloride was a Schwarz/Mann product; glucose, Mg acetate and Tris-HC1 were local products of analytical grade. Preparation and activity assay of adenylate kinase. The enzyme was prepared essentially according to Heft et al. [24] except that the elution of fraction III was done with 10 mM ADP and the Sephadex G-75 column was washed with 0.1 M imidazole-HC1 buffer (pH 7.0). The yield was usually about 60 mg of pure enzyme/kg rabbit skeletal muscle. The final preparations usually had a specific activity greater than 1600 units/mg and showed only one peak in both gel filtration and reverse-phase FPLC. One unit was defined as 1 /~mol ATP generated from ADP per min. The activity assay was made by following the reduction of NADP in a coupled enzyme solution with hexokinase and glucose-6-phosphate dehydrogenase. The reaction mixture contained 2.5 mM ADP, 2.1 mM Mg acetate, 6.7 mM glucose, 0.67 mM NADP, 20 units hexokinase and 10 units glucose-6-phosphate dehydrogenase in 50 mM Tris-HCl buffer (pH 8.1). Concentration of adenylate kinase was determined by absorption at 280 nm with an absorbance coefficient of elc m-°l~ = 0.52. Inactivation and reactivation. The inactivation of adenylate kinase was measured by first mixing 10 /~1 portions of the enzyme solution with 1 ml buffer previously heated to the denaturating temperature with a final enzyme concentration of 20 tzg/ml. Aliquots of 10/xl were withdrawn after different periods of incubation and added to 1 ml of the assay solution at 25°C, for measurements of the activity remaining. In some cases, inactivation was continuously monitored by following the substrate reaction at the denaturating temperatures using the kinetic method of Tsou [25], which has been shown to be applicable with a coupled en-

zyme assay system [26]. The final composition of the reaction mixture was as above, except that 100 units hexokinase, 50 units glucose-6-phosphate dehydrogenase and 0.2 /xg adenylate kinase were added. Hexokinase and glucose-6-phosphate dehydrogenase have been reported in the literature to be fairly thermostable [27] and found in the present study to have at least 60% the original activity under the required conditions at the highest temperatures employed. Furthermore, the activity measured for adenylate kinase under such conditions was found to be strictly linearly proportional to the amount of this enzyme present in the assay system. The inactivation rate constants obtained by both methods agreed satisfactorily. For the study of the reactivation of thermally-denatured adenylate kinase, the enzyme (1.0 mg/ml), denatured for 30 min under different temperatures was cooled at 4°C for 12 h. It was either diluted directly 40-fold at 4°C and left standing for 4 h before activity analysis or first with an equal volume of 6 M guanidine hydrochloride at 4°C, left standing for 4 h at this temperature and then diluted again 20-fold with buffer and left standing for another 4 h before analysis. The final dilution was 40-fold in both cases.

Conformational changes during unfolding and refolding. The thermal unfolding of adenylate kinase and refolding of the enzyme after cooling were studied by a number of physical methods as follows: A Jasco-500 A spectropolarimeter was used for CD measurements in the far ultraviolet region from 200250 nm. For the time course of unfolding, ellipticity was followed at a wavelength of 220 nm for changes in the helix content. Changes in fluorescence emission spectrum were measured with a Hitachi F-4010 spectrofluorometer with an excitation wavelength of 285 nm. To study the time-course, emission at 305 nm was followed as this enzyme contains Tyr but no Trp residues. Second derivative spectra were recorded with a Shimadzu UV 3000 spectrophotometer between 270 and 320 nm. For temperature difference spectra, the

TABLE I

Rate constants for the inactivation and conformational changes of adenylate kinase during thermal denaturation Temperature

Inactivation k

Denaturation k

(°C)

fast

CD

30 40 45 50 55

3.8 10.3 12.0 14.1 62

KJ

79

66

60

c

20

20

20

slow 0.24 0.42 0.88 1.38 1.5

a s

a s 0.85 1.05 1.7

2.2 5.8 7.2

Aggregation k Fluorescence

Scattering

a s

s 0.67 n 2.2 n

a s 2.1 2.56

n

4.7 10.8 n

Absorbance 0.85 2.17 n 3.7 n

2.07 n n n n

s 1.13 n 2.1 n

60 220

40

100

5

20

240

20

k, rate constant (s "1.103); a, no detectable change; s, change too small for rate constant determination; n, not measured; KJ, activation energy ( k J / m o l ) ; c, enzyme concentration (/xg/ml).

63 spectrum at 25°C was recorded and this was then subtracted from that for higher temperatures to obtain the difference spectra. The enzyme concentrations for the above measurements were summarized in Table I. Gel filtration was carried out with a Superose 12 H R 10/30 column on a Pharmacia FPLC apparatus. The enzyme denatured under different temperatures for 30 min was first centrifuged before chromatography. To follow the course of aggregation of the enzyme, light scattering was measured at a direction 90 ° to the incident light in the Hitachi F-4010 spectrofluorometer with the wavelength of both the incident and scattered light fixed at 488 nm. Alternatively, the light energy loss due to scattering was measured as apparent absorbance change at 488 nm in the Shimadzu UV 3000 spectrophotometer. At this wavelength scattering can be measured free from interferences due either to the intrinsic fluorescence or light absorption of the enzyme. The solution became noticeably turbid with an enzyme concentration of 0.2 m g / m l and a heating time of 30 rain at 50°C.

linear Arrhenius plot occurs below 15°C. The apparent activation energy for this reaction was found to be 38 kJ/mol.

Kinetics of thermal inactivation The time-course of thermal inactivation of adenylate kinase at different temperatures was studied and the results obtained at 30 and 50°C are shown in Fig. 2. The same results were obtained when the activity determinations were made at the denaturating temperature or after being cooled to 25°C, showing that no reactivation occurred when the inactivated enzyme was diluted in presence of the substrate at the lower temperature. The inset of Fig. 2 gives a Guggenheim plot for the data at 30°C, showing biphasic kinetics for the inactivation reaction. Similar biphasic inactivation kinetics were obtained at higher temperatures. The corresponding rate constants increase with increasing temperature as summarized in Table I. The activation energies obtained from Arrhenius plots were 79 and 66 kJ for the fast and slow phases, respectively, of the heat inactivation reaction.

Results

Far ultraviolet CD Arrhenius plot of the adenylate kinase catalyzed reaction Fig. 1 shows the Arrhenius plot of the temperature dependence of the activity of adenylate kinase. The enzyme is sufficiently stable, so that if initial rates only are followed the plot is linear below 50°C. Rapid thermal inactivation takes place above this temperature so that accurate initial rate measurements are no longer possible. As with most enzymes, a break of the

4.00

3.50

-5

Intrinsic fluorescence

3.00

E

'%

6, 0

_J

2.50

2.00 3.00

It is known from X-ray diffraction studies that adenylate kinase is an a-fl type of protein rich in both a-helix and fl-sheets [2] which is also shown by its CD spectrum typical for this type of molecule [28]. Fig. 3A shows that the two negative peaks at 220 and 208 nm decrease in magnitude with increasing temperature and disappear completely at 60°C. The thermal unfolding of the muscle enzyme was completely irreversible as followed by CD or other measurements. Semilogarithmic plots of the time courses of changes in mean residue eIlipticity at different temperatures (not shown) gave monophasic first-order kinetics and the rate constants thus obtained at two different concentrations are summarized in Table I.

"8 i 3.20

i 3.40

3.60

The emission spectra of the intrinsic fluorescence of adenylate kinase at different temperatures are shown in Fig. 3B. As this enzyme is devoid of Trp, its emission spectrum shows the typical peak of Tyr at 305 nm. The emission intensity decreases with increasing temperature with no noticeable shift of emission maximum. The time courses of emission changes at different temperatures also shows monophasic first order kinetics in semilogarithmic plots (not shown). The rate constants determined at two different protein concentrations are summarized in Table I.

1/T X 10`3 (K-1) Fig. 1. Arrhenius plot of the temperature dependence of the activity of adenylate kinase. The enzyme is sufficiently stable during activity assay so that the plot is linear below 50°C. Rapid thermal in~tctivation takes place above this temperature.

Second derivative absorption spectra Because of interference by light scattering due to heat aggregation of the enzyme, second derivative difference spectra of adenylate kinase at different tem-

64

>

0.80

0.1o

0.60 n~

w1

>O

0.40

~6 ~- 0.20

0 ~zx

,

A Z~, - -~ ~ A

0.00 0

20

40

60

80

100

120

to the incident light or by the energy loss of the transmitted light (apparent absorbance) at 488 nm, a wavelength at which no light absorption or fluorescence emission occurs for this enzyme. The courses of aggregation as followed by apparent absorption are shown in Fig. 4A at different temperatures and the rate constants as obtained by Guggenheim plots of results of both scattering or apparent absorption are summarized in Table I. Separate experiments by molecular sieve chromatography on Superose 12 also indicated aggregation for the thermally-denatured enzyme.

Time ( min )

Fig. 2. The course of thermal inactivation of adenylate kinase. The enzyme (0.02 mg/ml) was incubated at the indicated temperatures in 50 mM Tris-HCI buffer (pH 8.1) and aliquots were taken at different time intervals for activity assay at 25°C. The inset shows a Guggenheim plot of the data obtained at 30°C.

peratures were compared with that recorded at 25°C instead of direct difference spectra (not shown). The spectrum at 30°C shows only very low maxima and minima, indicating little conformation change at this temperature but the peaks at 279 and 286 nm and troughs at 284 and 292 nm increase markedly with further increase in temperature. As this enzyme contains only Tyr but no Trp residues, no peak at 297 nm was apparent, but the increase in A286 - A 2 8 4 from 30 to 55°C indicates increased exposure of the Tyr residueses at the higher temperatures [29].

Aggregation The relative state of aggregation of adenylate kinase was followed either by light scattering at a direction 90°

Effect of protein concentration on the unfolding and aggregation of adenylate kinase The effect of enzyme concentration on the extents of changes in CD and light scattering during thermal denaturation at different temperatures are shown in Fig. 4B and C, respectively. The secondary structure of this enzyme is more stable at a higher protein concentration (Fig. 4B). For the aggregation of adenylate kinase during thermal denaturation, the extents of aggregation in the lower temperature range appear to be less marked at a lower protein concentration as for CD changes but increased more markedly with increasing temperature and became more highly aggregated than that at the higher protein concentration at temperatures over 45°C (Fig. 4C). The rate constants of unfolding as followed by CD and fluorescence emission changes and the rate constant of aggregation as measured by light-scattering changes at different protein concentrations are summarized in Table I. It is to be noted that the rates of both 30

i

E

i

i

i 330

I 350

25"C

7 o

i

B

A 25

0

nO

E

20

-5

t,..) E

L)

O

tm

--10 I,,'3 I 0 X

15

o

"-I

u..

--15

10

-20

5

-25 190

~ 210

i 230

J 250

0 270

i 290

i 310

370

rim nm Fig. 3. (A), CD spectra of adenylate kinase at different temperatures. The enzyme (0.2 mg/ml) in 50 mM Tris-HCl buffer (pH 8.1) was allowed to stand at the indicated temperature for 30 min before CD measurements at the same temperature. (B), Fluorescence emission spectra of adenylate kinase at different temperatures. The enzyme (0.1 mg/ml) in 50 mM Tris-HCl buffer (pH 8.1) was allowed to stand at the indicated temperature for 30 min. It was then cooled at 4°C overnight before fluorescence measurements at 25°C with excitation wavelength of 285 nm.

65

0.05

Et-

A

Comparison of conformation and activity changes during thermal denaturation

60"C

0.04

OO 00 .,d-

0,03 50.C 0.02

~

0.01

30"C 0.00 0

i

i

i

i

i

i

5

10

15

20

25

30

35

Time ( min )

B

0.60

~>

0.20

[

[

0"0020

0

1.00 / C

|

0.40

#"

0.20

000

[ [ "

20

-

-

30

40

50

60

Temperature (°C) unfolding and aggregation during thermal denaturation of this enzyme are invariably higher at higher protein concentrations as examined by all the abovementioned methods.

Fig. 5 summarizes the effect of temperature on the changes in activity, CD, fluorescence, second derivative difference spectrum and aggregation of adenylate kinase after heating for 30 rain at a protein concentration of 20 /zg/ml. The rate constants of inactivation and conformational changes at different protein concentrations as followed with different physical methods are summarized in Table I which also summarizes the activation energies of the inactivation, unfolding and aggregation processes. It is clear from Fig. 5 that under the same protein concentration and heating time, inactivation occurs at a lower temperature than both conformational changes and aggregation as measured by all the methods employed. Moreover, the extents of unfolding and aggregation under different temperatures as obtained by different methods are not identical. The above results cannot be simply accounted for by a simple two-state model as reported for some other proteins [30-33]. From a comparison of the rate constants listed in Table I, the fast-phase inactivation rate constants at all the temperatures studied are about one order of magnitude greater than the corresponding rate constants for changes in aggregation (light scattering) and conformation (CD) measured at the same protein concentration. The rate constants of unfolding as measured by different methods are also not identical. The rates of CD changes at different temperatures are very close to those for the slow-phase inactivation, suggesting that the slow inactivation phase is accompanied by significant overall conformation changes. It is also to be noted that for the unfolding rates measured at two or more protein concentrations, higher unfolding and aggregation rates were invariably observed at the higher concentration. The difference in the rate constants would probably have been greater had these been

Fig. 4. (A), The course of relative changes in aggregation state of adenylate kinase at different temperatures. A concentrated solution of the enzyme (2 mg/ml) at 25°C was diluted into 50 mM Tris-HCI buffer at pH 8.1 thermostated at the indicated temperature to a final concentration of 0.02 mg/ml and the absorption at 488 nm due to energy loss was followed. (B,C) Thermal denaturation and aggregation of adenylate kinase at different protein concentrations. (B), Relative conformationchange as measured by mean residue ellipticity at 220 nm. Curves 1 and 2 are for two different protein concentrations of 220 and 22/zg/ml, respectively. The enzymewas in 50 mM Tris-HCl buffer (pH 8.1) for CD measurements at the indicated temperature. (C), Aggregation as monitored by light scattering (circles) and apparent absorbance (triangles) at 488 nm at the indicated temperature. Curves 1 and 2 are for two different protein concentrations, 20 and 5/zg/ml, respectively. The enzymewas in 50 mM Tris-HC! buffer (pH 8.1) for heat treatment and measurement at the indicated temperatures.

66 1.00

m

.

.

0.60 0.60

•~

H

0.40

W

0.20

~

0.00 20

30

40

50

60

T e m p e r a t u r e (*C)

Fig. 5. Comparison of inactivation and conformational changes of adenylate kinase at different temperatures. Final concentration of enzyme in 50 m M Tris-HCl buffer (pH 8.1) was 0.02 m g / m l and the heating time was 30 min in all cases during heat treatment. In order to obtain reliable results, cuvettes with a long light pathlength were used, 1 cm for CD and 10 cm for ultraviolet absorption measurements. The curves are: inactivation (1); light scattering (2); fluorescence m e a s u r e m e n t s made with excitation wavelength at 285 n m and emission wavelength 305 nm (3); CD m e a s u r e m e n t s at 220 n m (4) and second derivative difference spectra (5).

measured at the same protein concentration, as is the case for fluorescence emission changes.

Renaturation and reactivation of thermal-denatured adenylate kinase Unlike the reversible thermal inactivation and unfolding of the Escherichia coil enzyme as reported by Reinstein et al. (1990), the thermal denaturation and inactivation of muscle adenylate kinase is irreversible and accompanied by aggregation of the monomeric enzyme. Comparison of Fig. 6A and B shows that considerable ordered secondary structures but very little activity of the enzyme remained after heating at 1.00

I, l l l , . ~ , ,

A

8

n-

~

3

,

~

temperatures between 30-50°C for 30 min. After an incubation period of 12 h at 4°C and dilution of 40-fold with buffer, the thermally denatured enzyme showed slight further decrease in both activity (Fig. 6A) and ordered secondary structure as indicated by a decrease in magnitude of its mean residue ellipticity (Fig. 6B). However, whenthe enzyme denatured at temperatures up to 60°C for 30 min was first disaggregated by dilution with an equal volume of 6 M guanidine hydrochloride at 4°C for 4 h, followed by a further 20-fold dilution with buffer for another 4 h period, over 80% of both the activity and ordered secondary structure can be recovered. Similar results were obtained when refolding was followed by fluorescence measurements. Discussion

Inactivation of adenylate kinase precedes conformation changes during thermal denaturation As in the denaturation by guanidine hydrochloride or urea of some enzymes previously reported [1217,19,20], during the thermal denaturation of adenylate kinase, inactivation also precedes significant changes in ordered secondary structure, exposure of Tyr residues and aggregation state as followed by CD, fluorescence, second derivative difference ultraviolet spectroscopy and light scattering. As in the case of thermal denaturation of glyceraldehyde-3-phosphate dehydrogenase [23], the rapid decrease in enzyme activity before noticeable conformation changes cannot be ascribed to inhibition by binding of the denaturants as has been suggested for the inactivation of some enzymes before significant conformation change during guanidine hydrochloride or urea denaturation [22]. The above lends further support to the suggestion made previously that enzyme active sites are situated in limited molecular 1.00

,

0.80

0.60

0.60

0.40

0.40

1 2

13£

o

,

,

,

~k I

0.20

A

0.00

0.00 20

r~

[]

o.8o

0.20

8--

a

30

40

50

T e m p e r a t u r e (*C)

60

70

20

.30

40

50

60

70

Temperature(*C)

Fig. 6. (A) Reactivation of thermally-inactivated adenylate kinase. The enzyme (1.0 m g / m l ) in 50 m M Tris-HCl buffer (pH 8.1) was heated at the indicated temperature for 30 min. Curve 1, the thermally-inactivated enzyme was cooled directly for activity assay; curve 2, cooled to 4°C overnight and diluted 40-fold with buffer and left standing for another 4 h before assay; curve 3, diluted with an equal volume of 6 M guanidine hydrochloride, kept at 4°C for 4 h, then diluted 20-fold with buffer and kept at 4°C for another 4 h before activity assay. Activity assay was done at 25°C in all cases. Activity of the native enzyme was taken as 1. (B) Refolding of the thermally-inactivated adenylate kinase. T h e enzyme was heat treated as for Fig. 6A. Mean residue ellipticity at 220 n m was measured: Curve 1, directly after 30 rain heating; curve 2, after the heated sample was cooled to 4°C for 12 h and diluted 40-fold with buffer and kept another 4 h at 4°C; curve 3, the heated sample was first incubated in 3 M guanidine hydrochloride and then diluted as for curve 3 in Fig. 6A. All m e a s u r e m e n t s were m a d e at 25°C and the m e a n residue ellipticity of the native enzyme was taken as 1.

67 regions easily perturbed during denaturation and are consequently more flexible than the molecules as a whole [18].

Differences in the extents and rates of unfolding as measured by difference methods The extents of conformational changes during denaturation of some proteins have been reported to be independent on the methods employed for their measurement and unfolding and refolding are, therefore, considered to be all-or-none processes [30-33]. Recently, discrete intermediates during unfolding and refolding have been recognized for some other proteins [34-36]. It is clear from Fig. 5 and Table I that during thermal denaturation of adenylate kinase, not only inactivation occurs at lower temperatures and with faster rates than unfolding of the enzyme molecule under the same conditions, but also the extents and rates of unfolding during denaturation are not identical when measured by different physical methods, such as circular dichroism and intrinsic fluorescence (Table I). This was also observed for creatine kinase during denaturation by guanidine-HC1 and urea [12,13] and appears to be the case for many other enzymes [19,20]. All the above cannot be explained on the basis of a two-state model. Disaggregation and reactivation Both the thermal inactivation and unfolding are irreversible and incubation at 4°C of the partially-denatured enzyme results in further decrease in the activity and native conformation. However, 80% activity recovery is possible by first disaggregation of the denatured enzyme in 3 M guanidine hydrochloride followed by further dilution. It is possible that spontaneous disaggregation of the thermally-denatured enzyme does not occur on cooling and standing. Apparently, the heat aggregated enzyme dissociates only in guanidine-HCl and a substantial part of the disaggregated but still unfolded enzyme molecules then refolds spontaneously upon, removal of guanidine HCI to recover both the native conformation and the activity of the enzyme.

Acknowledgement The present investigation was supported in part by Grant No. 9388006 of the China Natural Science Foundation.

References 1 Noda, L. (1973) in Enzymes, 3rd FAn., Vol. 8, pp. 279-305, Academic Press, New York.

2 Dreusicke, D., Karplus, P.A. and Schulz, G.E. (1988) J. Mol. Biol. 199, 359-371. 3 Kuby, S.A., Mahowald, T.A. and Noltmann, E.A. (1962) Biochemistry 1, 748-762. 4 Yazawa, M. and Noda, L.H. (1976) J. Biol. Chem. 251, 3021-3026. 5 Smith, G.M. and Mildvan, A.S. (1982) Biochemistry 21, 61196123. 6 Shyy, Y.J., Tian, G. and Tsai, M.D. (1987) Biochemistry 26, 6411-6415. 7 Kim, H.J., Nishikawa, S., Tokutomy, W., Takenaka, H., Hamada, M., and Kuby, S.A. (1990) Biochemistry 29, 1107-1116. 8 Tsai, M.D. and Yan, H.G. (1991) Biochemistry 30, 6806-6818. 9 Fry, D.C., Kuby, S.A. and Mildvan, A.S. (1985) Biochemistry 24, 4680-4694. 10 Fry, D.C., Kuby, S.A. and Mildvan, ,~.S. (1987) Biochemistry 26, 1645-1655. 11 Liang, P., Phillips, G.N. and Glaser, M. (1991) Proteins Struct. Funct. Genet. 9, 28-36. 12 Yao, Q.Z., Tian, M., and Tsou, C.L. (1984) Biochemistry 23, 2740-2744. 13 Yao, Q.Z., Liang, S.J., Tian, M. and Tsou, C.L. (1985) Sci. Sin. 28B, 484-493. 14 Xie, G.F. and Tsou, C.L. (1987) Biochim. Biophys. Acta 911, 19-24. 15 Liang, S.J., Lin, Y.Z., Zhou, J.M., Tsou, C.L., Wu, P.Q. and Zhou, Z.K. (1990) Biochim. Biophys. Acta 1038, 240-246. 16 Liu, W. and Tsou, C.L. (1987) Biochim. Biophys. Acta 916, 455-464. 17 Ma, Y.Z., Tsou, C.L. (1991) Biochem. J. 277, 207-211. 18 Tsou, C.L. (1986) Trends Biochem. Sci. 11,427-429. 19 Kelly, S.M. and Price, N.C. (1991) Biochem. J. 275, 745-749. 20 West, S.M. Kelly, S.M. and Price, N.C. (1990) Biochim. Biophys. Acta 1037, 332-336. 21 Reistein, J., Vetter, I.R., Schlichting, I., R6sch, P., Wittinghofer, A. and Goody, R.S. (1990) Biochemistry 29, 7440-7450. 22 Creighton, T.E. (1990) Biochem. J. 270, 1-16. 23 Lin, Y.Z., Liang, S.J., Zhou, J.M., Tsou, C.L., Wu, P.Q. and Zhou, Z.Q. (1990) Biochim. Biophys. Acta 1038, 247-252. 24 Heil, A., Miiller, G., Noda, L., Pinder, T., Schirmer, H., Schirmer, I., and Zabern, I.V. (1974) Eur. J. Biochem. 43, 131-144. 25 Tsou, C.L. (1988) Adv. Enzymol. 61, 381-436. 26 Wang, Z.X. and Zhao, K.Y. (1992) Sci. Sin. 35B, 570-576. 27 White, T.K., Kim, J.Y. and Wilson, J.E. (1990) Arch. Biochem. Biophys. 276, 510-517. 28 Russell, P.J., Chinn, E., Williams, A., David-Dimarino, C., Taulane, J.P. and Lopez, R. (1990) J. Biol. Chem. 265, 1180411809. 29 Ragone, R., Colonna, G., Balestrieri, C., Servillo, L. and Trace, G. (1984). Biochemistry 23, 1871-1875. 30 Nojima, H., Ikai I., Oshima T. and Noda, H. (1977) J. Mol. Biol. 116, 429-442. 31 Lindsay, C.D. and Pain, R..H. (1990) Eur. J. Biochem. 192, 133-141. 32 Eftink, M.R., Ghiron, C.A., Kautz, R.A. and Fox, R.O. (1991) Biochemistry 30, 8945-8953. 33 Jackson, S.E. and Fersht, A.R. (1991) Biochemistry 30, 1042810435. 34 Horowitz, P.M. and Criscimagna, N.L. (1990) J. Biol. Chem. 265, 2576-2583. 35 Wedler, F.C. and McLean, M.A. (1991) Biochim. Biophys. Acta 1076, 161-163. 36 Herold, M. and Kirschner, K. (1990) Biochemistry 29, 1907-1913.