ARCHIVES

Vol.

OF BIOCHEMISTRY

287,

No.

1, May

AND

15, pp.

BIOPHYSICS

135-140,

1991

Reversible Inactivation of Rat Liver Inorganic Pyrophosphatase by Substrate and Its Analogs Irina

N. Smirnova

A. N. Belozersky

Received

July

and Alexander

Laboratory

30, 1990,

and

A. Baykovl

of Molecular

in revised

form

Biology

December

and

Bioorganic

Academic

Press,

University,

Moscow

119899,

USSR

Inc.

Inorganic pyrophosphatase (pyrophosphate hydrolase, EC 3.6.1.1) catalyzes hydrolysis and synthesis of PPi and is interesting for, at least, two reasons. First, pyrophosphatase is the simplest member of the group of the enzymes which reversibly transfer phosphoryl from a polyphosphate to water. The mechanism of such reactions constitutes one of the most intriguing problems of enzymology. Second, pyrophosphatase controls the concentration of PPi, which is a powerful regulator of numerous cell processes (1). Rat liver cells contain cytosolic and mitochondrial pyrophosphatase (2, 3). The predominant cytosolic pyrophosphatase was isolated in two laboratories and shown to be a dimer of identical subunits, 33-38 kDa each (35). Typically for all pyrophosphatases, it requires Mg2+ for activity. Kinetic studies of the hydrolytic reaction indicated that, in addition to the substrate Mg PPi, rat liver cytosolic pyrophosphatase binds three Mg2+ ions, of which whom

State

two are absolutely required for activity and one has a modulatory function (6). In this respect, this enzyme resembles better known microbial pyrophosphatases for which kinetic (7-16), binding (17-23), and X-ray (24) studies have provided vast evidence for the multiplicity of metal-binding sites. Comparative inhibition studies with a series of PPi analogs have revealed, however, significant differences in the organization of the active sites of animal and microbial pyrophosphatases (5). The data obtained in the present study show that rat liver pyrophosphatase is susceptible to reversible inhibition by PP;, a unique property of this pyrophosphatase, resulting from the existence of two interconvertible forms of the enzyme. This report also provides information about the rates of Mg2+ and PPi binding to and dissociation from the enzyme. MATERIALS

’ To

Moscow

6, 1990

Dissociation of Mg2+ from one of the two metal-binding sites whose occupancy is absolutely required for catalysis by rat liver inorganic pyrophosphatase is a slow reaction (r1,2 = 3 h). Polycarhoxylic Mg2+ complexons markedly accelerate this process due to their binding with Mg2+ on the enzyme. PPi, ATP and a number of diphosphonate analogs of PPi also hind with Mg2+ on the enzyme with concomitant decrease in enzyme activity by 75% hut do not release the hound Mg 2+. The resulting ternary complex rapidly (T~,~ of several seconds) dissociates upon dilution into substrate-free medium. PPi and imidodiphosphate, which are substrates for pyrophosphatase, decrease the rate of reactivation by at least two orders of magnitude. The results can be explained by existence of two interconvertible forms of the enzyme, of which one is inactive and is stabilized by substrate or its analogs. 8 1991

Chemistry,

correspondence

should

0003.9861/91 $3.00 Copyright 0 1991 by Academic Press, All rights of reproduction in any form

he addressed.

AND

METHODS

Highly purified rat liver cytosolic pyrophosphatase was obtained as described previously (5). Solutions of the enzyme (0.25-0.5 mg/ml) in 25 mM Tris-HCl buffer (pH 7.5 at 4°C) containing 1 mM MgSO,, 1 mM dithiothreitol, 0.01 mM EGTA,’ and 0.05% bovine serum albumin were kept frozen at -25°C. After being thawed, they were supplemented with fresh dithiothreit.01 to 5 mM and allowed to stand for at least 24 h at 4°C before use. Incubations of pyrophosphatase were done at 20-25°C in a total volume of 0.3 ml. Unless stated otherwise, the incubation medium contained 2.5 wg/ml enzyme, 0.05 M Tris-HCl (pH 7.2), 5 mM dithiothreitol, 0.01 mM EGTA, 0.05% albumin and MgSO, at indicated concentrations. The incubation was started by addition of PP,, its analog, or metalchelating agent. The stock solutions (50-100 mM) were adjusted to pH 7.2 with Tris. Aliquots of the incubation mixture (8-50 ~1) were withdrawn in time and added to 25 ml of 0.05 M TrissHCl buffer (pH 7.2) containing, if not otherwise indicated, 1 mM MgSO,, 0.14 mM PP,, and 0.01 mM EGTA at 25°C. Enzymatic activity was estimated from initial rates of Pi liberation monitored with an automatic analyzer (25) at a sensitivity of 2.5-25 PM P, per recorder scale. Product formation curves were virtually linear, indicating no significant reactivation during the

’ Abbreviations tetraacetate; raacetate.

EGTA,

used: CDTA, trans-1,2-diaminocyclohexane-N,N’ethylene glycol bis(@-aminoethyl ether)N’N’-tet-

135 Inc. reserved.

136

SMIRNOVA

6

AND

BAYKOV

100 80

80

60

c 60 .= .1 ;; 40 a

n

0

,\ I

Time

=

. 2

(h)

FIG. 1. Pyrophosphatase inactivation by PP, (0) and ATP (0). The inset shows the rapid phase of the inactivation by PP,. The ordinate is scaled logarithmically. (0) 4 mM PP,, 20 @4 MgSO,; (A) 4 mM PP,, 80 pM MgS04; (0) 1 mM PPi, 20 pM MgSO,; (0) 4 mM ATP, 20 pM MgSO,.

time of the activity assay (2 min). The activity of the enzyme preincubated without any additions did not change and was taken for 100%. For measurements of enzyme-bound magnesium, PP, and ATP, the concentration of pyrophosphatase in the incubation mixture was increased to 0.5 mg/ml. Centrifugal gel filtration was performed according to Penefsky (26). Centrifugation was carried out for 2 min at 400g at the bottom of the Sephadex gel, which was preequilibrated with 0.05 M Tris-HCI (pH 7.2) containing 1 mM dithiothreitol. Enzyme recovery was about 70% and was taken into account in subsequent calculations. The amount of Mg in the eluate was estimated using an AASlN atomic absorption photometer (Carl Zeiss, Jena). ‘*P was determined with a liquid scintillation counter.

RESULTS

Inactivation

0

,O 4

3 Time

(s)

of Pyrophosphatase by PPi and Its Analogs

Preincubation of pyrophosphatase with PPi or ATP in the presence of trace amounts of Mg2+ resulted in a marked loss of activity measured with 1 mM Mg2+ (Fig. 1). With PPi, activity rapidly dropped to a steady level while with ATP the process was clearly biphasic. The data for PP, (inset to Fig. 1) shows that the approach to the steady level of activity has a rI12 of about 5 s, which did not depend markedly on PPi and MgSO, concentrations over the ranges examined. Hydrolysis of PPi in the preincubation medium was negligible and could not explain the absence of the second phase with this effector. The degree of the inactivation depended on PP; and ATP concentrations (Fig. 2). With ATP, the effect shown in Fig. 2 refers to the activity loss associated with the rapid phase and was estimated by extrapolating the straight line for the slow phase in the semilogarithmic plot (Fig. 1) to zero time. The residual activity approached approximately 25% at increasing concentrations of both effecters. A number of PP; analogs were also found to inactivate pyrophosphatase in a manner similar to that observed with ATP. They included imidodiphosphate, which contains NH instead of 0 in the bridge position,

, 5

, IO

PP; or ATP

(mM)

FIG. 2. Degree of pyrophosphatase inactivation by PP, (0) and ATP (0) during the rapid phase as a function of effector concentration in the presence of 50 pM MgS04. The curves were constructed using Eq. [2] given in the Discussion section and the best-fit values of the parameters mentioned in the text. [Enzyme], 20 pg/ml.

and a series of diphosphonates of the structure 03PC(R)(R)-PO, where R and R’ are different substituents at the carbon atom (5). Pi exhibited quite a small effect, if any (data not shown). All preincubations reported above were performed in the presence of small amounts of Mg2+ coming from the stock solution of pyrophosphatase. The presence of this low level of Mg2+ did not affect the inactivation pattern but made the results more reproducible compared to those obtained with metal-depleted enzyme. At higher concentrations, Mg2+ affords protection against the inactivation, as demonstrated in Fig. 3 for aminomethanediphosphonate, a nonhydrolyzable substrate analog. The inactivation patterns for PPi, ATP, and diphosphonates did not change when enzyme concentration was varied in the range of 2.5400 pg/ml.

Time FIG. 3. The pyrophosphatase [MgS04] (mM): (V) 10.

(h)

effect of MgSO, concentration on the time-course of inactivation by 4 mM aminomethanediphosphonate. (0) 0.025; (0) 0.05; (A) 0.075; (A) 0.1; (0) 0.5; (m) 1;

REVERSIBLE

INACTIVATION

Time

OF

RAT

LIVER

(s)

FIG.

4. Reactivation of pyrophosphatase pretreated for 5 min with 4 PPi (0) or ATP (0). The inhibited enzyme was diluted 3000-fold into 0.05 M Tris-HCl (pH 7.2) containing 1 mM MgSO( and 0.01 mM EGTA, and the enzymatic reaction was initiated in different time intervals by addition of 0.14 mM PP,. The lines show the best fit for Eq. mM

[Il.

Reversibility

of the

Inactivation

by PP, and Its Analogs

In the experiments reported above, pyrophosphatase activity was measured by diluting the enzyme into the assay medium containing 140 PM PP;. If enzymatic activity was measured in the presence of only 1.4 yM PPi, which is close to the Michaelis constant for this substrate at 1 mM Mg2+ (5,6), no inactivation by PPi, ATP, or diphosphonates could be observed. Hence, PP; stabilizes the inactive form of the enzyme, which otherwise reactivates during the assay of activity. To check for the reversibility of the inactivation, the assay procedure was modified in the following way. The enzyme treated with ATP or PP; was diluted into the medium containing 1 mM MgS04 and no PPi and was incubated for different periods of time before addition of 140 /.LM PP;. Figure 4 shows that both PPi-inhibited enzyme and ATP-inhibited enzyme were restored to full activity but at different rates. These data can be described by the equation for a first-order reaction A -

A0

= (A,

-

A,)(

1 -

e- 0.693thz)

PYROPHOSPHATASE

BY

137

SUBSTRATE

tained while with ATP only the rapid phase of the inactivation could be reversed. The inactivation/reactivation cycle could be repeated several times. The enzyme treated with ATP and reactivated by dilution could be again inactivated by addition of ATP and further reactivated by addition of 5 mM MgC12. Further evidence for the stabilization of the inactive form of the enzyme by substrate was provided by experiments with imidodiphosphate, which is slowly hydrolyzed by rat liver pyrophosphatase. The addition of imidodiphosphate to the reactivation medium slowed down dramatically the restoration of enzymatic activity (Fig. 5). In the presence of 100 pM imidodiphosphate, the halftime for this reaction was 10 min, i.e., 240 times that in the absence of imidodiphosphate. Interestingly enough, aminomethanediphosphonate, which is not substrate for this enzyme (5), did not affect its reactivation (Fig. 5, inset). The above results provide strong evidence for formation of an enzyme-effector complex during the preincubation step. An attempt was made to isolate [“‘PIATP- or PPienzyme complex formed with labeled effecters by centrifugal gel filtration in the presence of imidodiphosphate, which, as shown above, inhibits the reactivation. The enzyme was first inactivated by 4 mM 32P-labeled ATP or PPi, then supplemented with 0.1 mM imidodiphosphate and 1 mM MgCl, and separated from unbound effecters by centrifugal gel filtration. The column buffer also contained imidodiphosphate and MgC12 to minimize reactivation during the separation. After this procedure, about 50% of the enzyme remained inactive, but no 32P was

(1)

in which Ao, A, and A,, refer to the activity at times 0, t, and infinity, respectively. The values of 71/Zfor the ATPand PPi-inhibited enzyme were found to be 2.5 and 8 s, respectively. The rate of the reactivation of the PPi-inhibited enzyme did not change when Mg2+ concentration in the medium was varied from 0.1 to 1 mM. If, however, no Mg2+ was present at all, the half-time for the reactivation dropped to about 1 s. The degree of activity recovery depended on the nature of the effector used to inactivate the enzyme. With PPi, full recovery was ob-

0

‘-1 ow IO

Time FIG.

20

(mln)

The effect of PP, analogs on the reactivation of pyrophosphatase preincubated for 5 min with 4 mM ATP. The inset shows reactivation on a different time scale. For details, see the legend to 4. The reactivation medium contained the following PPi analogs: none; (a) 20 pM imidodiphosphate; (0) 100 pM imidodiphosphate; 100 pM aminomethanediphosphonate; (m) 200 pM aminomethanediphosphonate. The lines show the best fit for Eq. [I]. 5.

the Fig. (0) (v)

138

SMIRNOVA

100w R

,”

8Ok\

2o0

I IO

-Y

I 20 Time

, 30

(min)

FIG. 6. Pyrophosphatase inactivation by CDTA and EGTA in the presence of 25 FM MgSO,. Zero time corresponds to the moments of CDTA or EGTA additions. The first point is for 15 s incubation time. The ordinate is scaled logarithmically. (V) 4 mM EGTA; (0) 0.05 mM CDTA; (A)O.lmM CDTA;(Cl)4 mM CDTA;(m)4 mM CDTA added 30 s after 4 mM PP;.

AND

BAYKOV

with the same buffer but containing 1 mM MgC12 instead of CDTA, its activity was restored with a half-time of approximately 8 min, i.e., much slower than for the enzyme inhibited by PPi or its analogs. Some reactivation was also observed during the assay of activity. Preincubation of the enzyme with 4 mM PPi did not affect the pattern of inactivation with 4 mM CDTA added 30 s later (Fig. 6). Since preincubation with PPi alone results in deep inactivation (Fig. l), this finding may mean that CDTA initially restored the activity by displacing PPi from the enzyme. At lower concentrations of CDTA (0.05-0.1 mM), the effect of PP, was only partially reversed. The inactivation patterns were identical whether PP; was added 30 s before or after CDTA (data not shown). This indicated that the equilibrium of the enzyme-effector interactions was attained at each moment of the inactivation process. Approximately half of the effect of 4 mM PPi was reversed in the presence of 0.05 mM CDTA, indicating that CDTA binds to the enzyme 80 times as tightly as PP; does in terms of dissociation constant. Measurement of the Magnesium Content of Pyrophosphatase

found associated with the enzyme. These results may mean that added imidodiphosphate exchanges with bound effecters on the enzyme. No bound radioactivity was found without imidodiphosphate either since the reactivation is too fast on the time scale of the gel filtration. The Effect of Metal-Ion Chelators on Pyrophosphatase Incubation of the enzyme with CDTA resulted in a slow loss of activity (Fig. 6). The rate of this process increased with increasing CDTA concentration. The process was, at least, biphasic as indicated by nonlinearity of the semilogarithmic plot. Similar effects were observed with EDTA (data not shown). In contrast, EGTA, which has a low affinity to Mg2+ (27), did not affect pyrophosphatase activity under identical conditions (Fig. 5), supporting the notion that the inactivation involves formation of a metal-chelator bond on the enzyme. The degree of the inactivation by CDTA was identical whether activity was measured with 140 or 1.4 PM PPi, indicating that the substrate does not stabilize this complex. Inactivation of pyrophosphatase could be effected also by a 600-fold dilution of the enzyme preincubated with 1 mM MgClz into a Mg’+-deficient medium (data not shown). In this case, the inactivation proceeded much slower (71/z = 3 h), indicating that the effects of CDTA shown in Fig. 6 are due to direct association with pyrophosphatase-bound Mg2+, rather than to sequestration of Mg2+ in solution. The inactivation caused by CDTA could be completely reversed. When the enzyme whose activity dropped to 21% after treatment with 3 mM CDTA was diluted fivefold

The effects described above suggest the presence of enzyme-bound Mg2’, which could be removed upon incubation with various effecters. Therefore, direct measurements of Mg in rat liver pyrophosphatase were performed using atomic absorption technique. The enzyme was preincubated with 50 PM MgC12 alone or in the presence of various effecters shown in Table I and subjected to the rapid gel filtration into a medium lacking Mg2’. The molar content of Mg in the eluted protein was calculated using its molecular mass of 70 kDa (5). The concentration of the protein was measured with a dye-binding procedure (28). It is seen that the enzyme preincubated with Mg2’ alone contains 1.4 mol Mg/mol and this value did not change significantly when the enzyme was incubated with PPi or ATP. In contrast, prolonged incubation with

TABLE

I

The Amount of Bound Magnesium in the Enzyme Preincubated with 50 PM MgCl, in the Presenceof Various Effecters as Measuredby the Gel Filtration Procedure Effector None f’f’, (4 mM) ATP (4 mM) ATP (4 mM) CDTA (4 mM) CDTA (4 mM) CDTA (4 mM)

Incubation (min) 0.75 0.75 180 0.75 20 180

time

Mg content (mol/mol) 1.4 1.3 1.3 1.1 1.2 0.45 0.1

f f -+ f + f k

0.1 0.2 0.25 0.2 0.1 0.05 0.05

REVERSIBLE

INACTIVATION

OF

RAT

LIVER

PYROPHOSPHATASE

BY

SUBSTRATE

139

effector complex is supported by the existence of significant residual activity even at saturating concentrations of all effecters. Also, a mechanism with only one enzymeeffector complex would be difficult to reconcile with the ability of PP;, a substrate which is rapidly hydrolyzed in the presence of Mg2+, to keep the enzyme inactive during DISCUSSION the assay of activity. PPi, ATP, and diphosphonates form relatively stable Most of the above data can be explained by a similar complexes with Mg2+ (27) and it is not unreasonable to mechanism in which slow isomerization of EM is followed expect that their effects on pyrophosphatase activity are by rapid binding of L. For this mechanism, however, encaused by removal of protein-bound Mg’+, insofar as it zyme activity is expected to approach zero at saturating rebinds slowly during the assay of activity (6). Although effector concentrations and the reactivation rate is not this mechanism is clearly operative with CDTA, the ef- expected to depend on effector nature. fects of PPi and its analogs are more complex and suggest Numerical values of the parameters shown in Scheme formation of ternary complexes with two forms of the 1 can be estimated in the following way. Relative residual enzyme as shown in Scheme 1: activity corresponding to the first inactivation step, a, is proportional to the ratio ([EM] + [EML])/([EM] + [EML] + [E*ML]) and is given by CDTA led to a disappearance of Mg in pyrophosphatase. Controls run in the absence of the enzyme indicated that the gel filtration procedure used could remove all Mg2+ not bound to the protein.

Here, E, M, and L are enzyme, metal, and PPi or its analog, respectively. According to the proposed mechanism, the interaction between EM and L involves two reversible steps: rapid binding of L and slow isomerization of EML. For ATP and diphosphonates, the latter step is followed by irreversible inactivation with the rate constant h3. For PP;, h3 = 0. EML rapidly dissociates when diluted, yielding active EM, which can bind and convert PP,. The fate of E * ML depends on substrate concentration in the dilution medium. When present at high levels, the substrates exchange with L on the enzyme before the backward isomerization (k-J can occur and the enzyme remains inactive. If substrate concentration is low, a significant fraction of EML dissociates to give active EM, enabling thus further conversion of E * ML into EML. In the presence of 1.4 pM PPi, the reactivation is complete within the initial phase of activity assay, which is not resolved with the method used (25). This could explain why no inactivation was observed at low PPi concentration in the assay medium. The second inactivation step observed with ATP and diphosphonates (k3) could not be reversed by dilution into effector-free medium and may be related to metal release from the enzyme. The metal content of pyrophosphatase did decrease on a 180-min incubation with ATP (Table I), but the effect was within the experimental error. Clearly, further work is needed to determine the mechanism of this reaction. The proposed mechanism of pyrophosphatase inactivation is supported by several lines of evidence. Inactivation rate is independent of effector concentration (Fig. l), consistent with this mechanism for [L] > K1 (see below). Formation of inactive pyrophosphatase-effector complex is also implicated by competition between PPi and CDTA (Fig. 6) and by the dependence of the reactivation rate on the nature of the effector (Fig. 4). The existence of the isomeric forms of the pyrophosphatase-

a=

K, + WLI (K2 + UK1

K1 +

[21

Fitting Eq. [2] to the data shown in Fig. 2 yielded the following values: K, = 0.26 mM, K2 = 0.31 (PPi) and K1 = 1.1 mM, K2 = 0.41 (ATP). The value of k2 for PPi is 8.3 mini’ (Fig. 1). Therefore, kp2 can be calculated to be 2.8 min-‘. From the data in Fig. 4, one can estimate k_, to be 5.2 mini’ in the presence of 1 mM Mg2+, which agrees reasonably with the above value indicating thus minor effects of Mgzf concentration on kp2. If, however, no Mg2+ was present at all, km2increased to 40 mini. The value of k3 is 0.006 mini’ for ATP as calculated from the slope of the line referring to the second inactivation step in Fig. 1. The protective effect of Mg2+ against inactivation (Fig. 3) can be explained by increase in k2 since neither K, (5) nor km2seem to change markedly at increasing Mg2+ concentration. In other words, Mg2+ causes a shift of the equilibrium of the isomerization reaction to the left. The effect can be explained by filling up another Mg2+-binding site (not shown in Scheme 1) whose existence has been indicated by kinetic studies of PPi hydrolysis (6). The high value of K2 probably also explains the lack of the rapid inactivation step with CDTA since the data shown in Fig. 6 clearly indicate that this effector binds to the enzyme before it removes magnesium from it. Our data provide direct evidence for the presence of bound magnesium in rat liver pyrophosphatase. The metal dissociates slowly even in the presence of strong metalion chelator, and on this basis this pyrophosphatase can be classified as a metalloenzyme. Within the precision of enzyme concentration measurement, the amount of bound magnesium is about 1 mol/mol subunit. No tightly bound magnesium was found in yeast and Escherichia coli pyrophosphatases [Ref. (19) and unpublished data of Smirnova, I. N., and Baykov, A. A.], and these enzymes are

140

SMIRNOVA

not inactivated by incubation with PP; and its analogs. These findings provide indirect support for the proposed mechanism of rat liver enzyme inactivation. Krishnan and Gnanam reported recently that chloroplast pyrophosphatase is inactivated by EDTA in a manner similar to that observed with the rat liver enzyme and is also affected by PPi (31). It would be interesting to determine the rates of Mg2+ binding to and dissociating from this enzyme. Earlier kinetic studies of rat liver pyrophosphatase showed that Mg 2t addition to one of the two metal-binding sites whose occupancy is required for activity is a slow reaction. The dissociation constant for the resulting complex was determined to be 28 ? 10 PM (6). Stability of EM can be also estimated from the data obtained in this work. The half-time for the dissociation of EM in the absence of any chelator is 3 h, which corresponds to the rate constant of 0.23 h-‘. The value of the rate constant for the reverse reaction can be calculated from the rate of CDTA-treated enzyme reactivation to be 5200 M-’ h-‘. The dissociation constant for EM is thus 44 PM, in fair agreement with the value derived from the steady-state kinetics of PP; hydrolysis. On the basis of these considerations, we conclude that the tightly bound magnesium is required for catalytic activity of this enzyme.

AND

BAYKOV 7. Moe,

0. A., and Butler,

8. Rapoport, Rapoport, 9. Baykov,

We thank Dr. 0. V. Krokhin for magnesium analyses with atomic absorption spectrometer and Dr. P. Enriquez for critical reading of the manuscript.

REFERENCES 1. Stetten, D. (1960) Amer. J. Med. 28, 867-870. 2. Irie, M., Yabuta, A., Kimura, K., Shindo, Y., and Tomita, K. (1970) J. Biochem. 67, 47-58. 3. Kostenko, E. B., Smirnova, I. N., Baykov, A. A., and Avaeva, S. M. (1983) Biokhimlya 48, 707-713. 4. Yoshida, C., Shah, H., and Weinhouse, S. (1982) Cancer Res. 42, 3526-3531. 5. Smirnova, I. N., Kudryavtseva, N. A., Komissarenko, S. V., Tarusova, N. B., and Baykov, A. A. (1988) Arch. Biochem. Biophys. 267, 280-284. 6. Unguryte, A., Smirnova, I. N., and Baykov, A. A. (1989) Arch. Biochem. Biophys. 273,292-300.

J. Biol.

A. A., and Avaeva,

Chem.

247,7308-7314.

J. G., Heitmann, 26,

S. M. (1973)

P., and

237-246.

Eur. J. Biochem.

136-

32,

142. 10. Braga, E., and Avaeva, 49, 528535.

S. M. (1972) Biochem.

11. Kasho, V. N., Duzhenko, 43,50-57. 12. Unemoto, Acta

Biophys.

V. S., and Avaeva,

T., Tanaka,

M., and Hyashi,

Res. Commun.

S. M. (1978)

M. (1973)

Biokhimiya

Biochim.

Biophys.

327,490-500.

13. Schreier,

E., and Hohne,

14. Knight,

W. B., Fitts,

chemistry

20,

15. Kasho, 321.

W. E. (1978)

16. Kurilova, Khim.

17. Ridlington, 7303-7307.

Lett.

S. M. (1984)

Znt.

S. M.

L. G. (1972)

A. A., Tam-Villoslado, Biophys.

20. Cooperman, 21. Welsh,

Acta

K. M., Armitage, 22, 1046-1054.

22. Knight, franca,

J. Biol.

Bioorg.

Chem.

247,

Avaeva,

S. M.

S. (1979)

228-238.

A., Springs,

B., and Hamm,

I. M., and Cooperman,

W. B., Dunaway-Mariano, J. J. (1984) J. Biol. Chem.

23. Kurilova, S. A., Bogdanova, S. M. (1984) Bioorg. Khim.

(1983)

P., and Rapoport,

J. J., and

569,

B. S., Panackal, 20, 6051-6060.

Biochemistry

Bio-

16, 315-

J. Biochem.

18. Rapoport, T. A., Hohne, W. E., Heitmann, (1973) Eur. J. Biochem. 33, 341-347. Biochim.

D. (1981)

T. I., and Avaeva,

J. W., and Butler,

19. Baykov,

SO, 93-96.

4079-4086.

S. A., Nazarova, 1032-1039.

9,

FEBS

S. N., and Dunaway-Mariano,

V. N., and Avaeva,

chemistry

ACKNOWLEDGMENTS

L. G. (1972)

T. A., Hohne, W. E., Reich, S. M. (1972) Eur. J. Biochem.

B. S. (1983)

D., Ransom, 259,

D. J. (1981) Bio-

S. C., and Villa-

2886-2895.

A. V., Nazarova,

T. I., and

Avaeva,

10, 115331160.

24. Terzian, S. S., Voronova, A. A., Smirnova, E. A., Kuranova, I. P., Nekrasov, Ju. V., Harutyunyan, E. E., Vainshtein, B. K., Hohne, W., and Hansen, G. (1984) Bioorg. Khim. 10, 1469-1481. 25. Baykov,

A. A., and Avaeva,

S. M. (1981)

Anal.

Biochem.

26. Penefsky, H. S. (1977) J. Biol. Chem. 252,2892-2899. 27. Sillen, L. G., and Martell, A. A. (1964) Stability Constants Ion Complexes, The Chemical Society, London. 28. Bradford, 29. Volk, (1982)

30. Baykov, Arch.

M. M. (1976)

Anal.

Biochem.

72,

A. A., Duzhenko,

Eur.

125, 215-220. A. R., Kasho, V. N., and Avaeva,

A. A., Pavlov, Biochem.

Biophys.

of Metal-

248-254.

S. E., Baykov, J. Biochem.

116,1-4.

V. S., and

Avaeva,

S. M.

S. M. (1989)

273,301-308.

V. A., and Gnanam, 31. Krishnan, 260, 277-284.

A. (1988)

Arch.

Biochem.

Biophys.