Keith Tornheim. sibility, the kinetics of activation of rat skeletal muscle

Transcription

1 THE JOURNAL OF BIOLOGICAL CHEMISTRY Vol. 260, No. 13, Issue of July 5, pp , by The American Society of Biolowal Chemists, Inc. Printed in U. S. A. Activation of Muscle Phosphofructokinase by Fructose 2,6-Bisphosphate and Fructose 1,6-Bisphosphate Is Differently Affected by Other Regulatory Metabolites* (Received for publication, November 2, 1984) Keith Tornheim From the Department of Biochemistry and the Division of Diabetes and Metabolism, Boston University School of Medicine, Boston, Massachusetts Fructose-2,6-Pz and fructose- 1,6-Pz are strong ac- has been viewed as a relatively poor analogue of fructose-2,6- tivators of musclephosphofructokinase.theyhave P2 and of little physiological importance. been shown to be competitive in binding studies, and it Although muscle and liver isozymes of phosphofructokinase is generally thought that they affect the physical and have similar regulatory properties, they differ in the relative catalytic properties of the enzyme in the same manner. potency of various activators and inhibitors (1, 2, 10). Fruc- However, there are indications in published data that tose-1,6-pz is a much stronger activator of muscle phosphothe effects of the two fructose bisphosphates on phos- fructokinase than of the liver isozyme (11). Furthermore, the phofructokinase are not identical. To examine this pos- levels of fructose-2,6-p2 are lower in muscle than in liver (121, sibility, the kinetics of activation of rat skeletal muscle and the levels of fructose-1,6-p2 are somewhat higher. Thus, phosphofructokinase by the two fructose bisphosphates were compared in the presence of other regulatory metabolites. Citrate greatly increased the of the enzyme for fructose-2,6-p,, with little effect on the maximum activation. In contrast, citrate greatly decreased the maximum activation by fructose- l,6-pz, with only a small effect on the Changes in the concentrations of the inhibitor ATP or the activator AMP similarly altered the Ko.6for fructose-2,6-pz, but altered the maximum activation by fructose-1,6-pz. Finally, when fructose-1,6-pz was added in the presence of a given concentration of fructose-2,6-p,, phosphofructokinase activity was decreased if the activation by fructose-2,6-pz alone was greaterthanthe maximum activation by fructose- 1,6-Pz alone. These results are consistent with competition of the two fructose bisphosphates for the same binding site, but indicate that the conformational changes produced by their binding are different. The reaction catalyzed by phosphofructokinase is the primary control point of the glycolytic pathway, and the activity of the enzyme is affected by a variety of metabolites (1, 2). Inhibition by ATP and activation by AMP make the enzyme responsive to the energy state of the cell, and inhibition by citrate signals the availability of alternative fuels. One somewhat puzzling property is activation of the enzyme by its product, fructose-1,6-p,, which can lead to oscillatory behavior of glycolysis in muscle extracts under certain conditions (3, 4). In recent years, a new regulatory metabolite has been discovered fructose-2,6-p2 (5-9). Decreases and increases in the hepatic concentration of this potent activator of phosphofructokinase are involved in mediating the effects of glucagon and glucose administration on hepatic glycolysis. Since the affinity of liver phosphofructokinase for fructose-2,6-p2 is about 1000 times its affinity for the 1,6-isomer, the latter * This work was supported by United States Public Health Service Grant AM The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact fructose-1,6-p2 is more likely to play a role in the physiological regulation of phosphofructokinase in muscle. The two fructose bisphosphates compete with each other for binding to muscle phosphofructokinase (11, 13) and have similar effects in promoting aggregation of the enzyme and on thiol reactivity and maximum activation of in vitro phosphorylation by the catalytic subunit of cyclic AMP-dependent protein kinase (11). These findings have been taken as evidence that the two fructose bisphosphates bind to the same site on the enzyme and that binding of either ligand produces the same conformational change. However, the effect of fructose-1,6-pz in promoting in vitro phosphorylation of the enzyme is reduced in the presence oflow concentrations of citrate, whereas the effectiveness of fructose-2,6-p, is much less sensitive to citrate (11). Examination of kinetic data from various sources also suggests possible differences in activation of the enzyme by fructose-2,6-p2 and fructose-1,6-p2. Thus, we have shown that the maximum activation of muscle phosphofructokinase by fructose-1,6-pz is altered by changes in the concentrations of AMP, ATP, and citrate (14), whereas Van Schaftingen et al. (15) have reported that AMP and ATP affect the K0.5(concentration of activator giving half-maximal activation) of the liver isozyme for fructose-2,6-p,, rather than the maximum activation. Uyeda et al. (16) have shown that fructose-2,6-p2 can relieve citrate inhibition of liver phosphofructokinase. Although not discussed, the data in their published figure could be redrawn to indicate an effect of citrate on the K0.5 of fructose-2,6-pz, rather than on the maximum activation. The latter authors also stated that they observed similar results with the muscle enzyme, but did not show the data. Finally, Kemp and Foe (17) pointed out that the maximum activation of muscle phosphofructokinase by glucose-1,6-p2 (which acts similarly to fructose-1,6-p2 but binds less strongly to the enzyme) was less than the maximum activation by fructose-2,6-pz when the two bisphosphates were compared in the presence of 1 mm citrate. Data obtained in the absence of citrate were not presented. In light of these findings, the present study was designed to compare the kinetics of activation of muscle phosphofructokinase by the two fructose bisphosphates under identical conditions and to examine interactions with other regulatory metabolites and

2 7986 Activation of with each other. The results confirm the suspected qualitative differences between the two fructose bisphosphates and therefore indicate that the regulation of phosphofructokinase cannot be viewed in terms of a simple, two-state conformation model. Phosphofructokinase by Fructose Bisphosphates -1 Kitratel mm 71 EXPERIMENTAL PROCEDURES Phosphofructokinase was partially purified from rat skeletal muscle as described previously (14). except that the tissue was broken up using a Polytron homogenizer. Reaction mixtures for kinetic studies contained 0.1 mm fructose-6-p, 0.33 mm glucose-6-p, 25 mm imidazole-hci buffer, ph 7.0, 8 mm M&Iz, 50 mm KCI, 1 mm EDTA, 0.5 mm phosphoenolpyruvate, 60 pm NADH, 1 mm triethanolamine, 6 mm orthophosphate, 20 pg/ml pyruvate kinase, 20 pg/ml lactate dehydrogenase, 12 pg/ml aldolase, 5 pg/ml phosphoglucose isomerase, and various concentrations of fncbse-2,6-pz, fructose-1,6-pz, ATP, AMP, and citrate. When 10 mm ATP was used, MgClz and NADH concentrations were increased to 16 mm and 240pM, respectively. The reaction mixture was preincubated in a water-jacketed cuvette holder at 30 C for about 5 min before addition of phosphofructokinase (1.6 pg of protein/ml, or about 0.04 unit/ml). Reaction rates were measured using a Hewlett-Packard model 8450 spectrophotometer system set to reada-46minusa3-. Equilibration of fructosel,6-p~ in the aldolase reaction was used to buffer the concentrations of fnctose-1,6-p2 (14), and equilibrium concentrations were determined as previously described. Similarly, the presence of glucose-6-p and phosphoglucose isomerase stabilized the fructose-6-p concentration and thus helped maintain linear rates. Fructose-2,6-P2 was purchased from Sigma. Since this compound is very acid-labile, stock solutions and dilutions were made in 5 mm triethanolamine buffer, ph 8. Fructose-2,6-Pz was assayed enzymatically as fructose-6-p following acid hydrolysis (ph 2-3 at room temperature for 10 min). Fructose-6-P content of the preparation before acid hydrolysis was about 0.5%. There was no detectible contamination with frucbse-1,6-p* (<0.2%). Fructose-1,6-P2 was purchased from Boehringer Mannheim. It contained 0.2% fructose-6-p, which was not increased by acid treatment to hydrolyze fructose-2,6-p2. Other biochemicals were obtained from Boehringer Mannheim or Sigma, except that sodium citrate was from Mallinckrodt Chemical Works. Auxiliary enzymes were obtained from Boehringer Mannheim. Aldolase and phosphoglucose isomerase were crystalline suspensions in ammonium sulfate. The ammonium sulfate was removed by gel filtration with an elution buffer of 20 mm potassium phosphate, ph 7. Lactate dehydrogenase and pyruvate kinase were obtained as solutions in 50% glycerol and were diluted in phosphate and imidazole buffer, respectively. Male Sprague-Dawley rats were obtained from Charles River Breeding Laboratories, Inc. RESULTS AND DISCUSSION In the presence of 1 mm ATP, 0.1 mm fructose-6-p, and 10 p~ AMP, muscle phosphofructokinase was strongly activated by submicromolar concentrations of fructose-2,6-p2. The under these conditions was about 0.3 p~ (Fig. 1). Addition of citrate in the physiological concentration range greatly decreased the affinity of the enzyme for fructose-2,6-p2. For example, at 0.5 r n citrate, ~ the KO.5 for fructose-2,6-p2 was between 1 and 2 p ~ In. comparison with the large effects on the affinity for f~ctose-2,6-p2, addition of citrate caused little if any decrease in the maximum level of activation. Under these same conditions, fructose-1,6-p2 also strongly activated muscle phosphofructokinase. In the absence of citrate, the Ko.5f0r fructose-1,6-p2 was about 10-fold higher than for fructose-2,6-p2 (Fig. 2), in agreement with earlier reports (11). However, in contrast to the results for fructose-2,6-p2, citrate had a relatively small effect on the K0.5 for fructosel,6-p2, but it markedly decreased the activation at saturating levels of fructose-1,6-p2. Similar qualitative differences in effects on the saturation curves of fructose-2,6-p2 and fructose-1,6-p2 were observed with the inhibitor ATP and the activator AMP. Increasing the ATP concentration or decreasing the AMP concentration r*.r( E \ E 1 v > [Fructose 2.6-bisphosphatel (pm) FIG. 1. Effect of citrate on the activation of muscle phosphofructokinase by fructose-2,0-p2. The reaction mixtures contained 1 mm ATP, 10 PM AMP, and the indicated concentrations of citrate. Other conditions were as given in the text. I Kitratel mm 4- P 3-.-I E \ x 1 > 2l I II IO [Fructose 1.6-bisphosphatel (pm) FIG. 2. Effect of citrate on the activation of muscle phosphofructokinase by fructose-1,6-p2. Conditions were as described for Fig. l. increased the K0.6 for fructose-2,6-p2 with little, if any, effect on the maximum activation (Fig. 3). For fructose-l,6-p2 tested under identical conditions, such changes in ATP or AMP concentrations primarily affected the maximum activation rather than the (Fig. 4). Evidence of competition in binding studies has suggested that both fructose bisphosphates bind to the same site(s) on the enzyme (11,13). However, if they activate the enzyme by binding at the same site, it is somewhat unexpected that there are such striking qualitative differences in the effects of other regulatory metabolites on the kinetics of activation. Experimenta were therefore performed to assess the interaction of fructose-2,6-p2 and fructose-1,6-p2 in activating phosphofructokinase. As shown in Fig. 5, when fructose-2,6-p2 was present at a low concentration, micromolar concentrations of fructose-1,6-p2 gave further activation. However, when fructose-

3 Activation of Phosphofructokinuse by Fructose Bisphosphates 7987 I O x [Fructose 2.6-bisphosphatel (pm) FIG. 3. Effect of changes in AMP or ATP concentrations on the activation of phosphofructokinase by fructose-2,6-p2. Conditions were as for Fig. I, except that citrate was not present and AMP and ATP concentrations were varied as indicated [Fructose 1.6-biephosphatel (pm) FIG. 5. Competition of fructose-l,6-pp and fructose-2,6-p, in activating phosphofructokinase. Conditions were as described for Fig. 1 in the absence of citrate. The AMP concentration was reduced to 3 p~ in order to decrease the maximum activation by fructose-1,6-p2 and thus accentuate the apparent inhibition by fructose-1,6-p2 in the presence of fmctose-2,6-p2. A L CATPI [AMPI mm pm A a0 io $0 io io io [Fructose 1.6-bisphosphatel (pm) FIG. 4. Effect of changes in AMP or ATP concentrations on the activation ofphosphofructokinase by fructose-1,6-p*. Conditions were as described for Fig. 3. 2,6-Pz was present at a concentration that activated the site but that their apparent enzyme substantially more than did saturatingconcentrations of fructose-1,6-pz, then fructose-1,6-pz appeared to be inhibitory. Phosphofructokinase activity at saturating concentrations of fructose-1,6-pz was not affected by fructose-2,6-pz. This is in contrast to citrate, ATP, and AMP, all of which altered the V, for fructose-1,6-p2. These results are consistent with competition of the two fructose bisphosphates for the same binding site. Most activators and inhibitors of phosphofructokinase, such as AMP, ATP, citrate, fructose-1,6-p2, and fructose-2,6- P2, act primarily by increasing or decreasing the affinity of the enzyme for its substrate fructose-6-p (2,8,14). a In simple model of enzyme regulation such as that of Monod et al. (18). the enzyme would be viewed as existing in two conformational states: an active state, characterized by high affinity for A 0 3 fructose-6-p and high affinity for activators as well, and an inactive or less active state, characterized by low affinity for fructose-6-p and high affinity for allosteric inhibitors. Pettigrew and Frieden (19) proposed such a two-state model for muscle phosphofructokinase, in which a difference in protonation would account for the known ph dependence of the regulatory properties. The kinetic experiments presented here are inconsistent with such a simple model and indicate that the active conformation with fructose-2,6-p2 bound is probably not the same as that with fructose-1,6-pz bound. The observation that citrate and ATP decrease the V, for fructose-1,6-p2 but not for fructose-2,6-p2 implies that these inhibitors can bind to the enzyme when fructose-1,6-p, is bound, but not when fructose-2,6-pz is bound; in other words, they are competitive with fructose-2,6-p2 but noncompetitive with fructose-1,6-p2. It is unlikely that this difference could be due to direct steric blocking by fructose-2,6-p2 (but not fructose-1,6-pz) of both ATP and citrate binding, since ATP and citrate act synergistically and therefore do not themselves bind at the same ok overlapping sites. Since there must be two active conformational states, it is of course possible that fructose-2,6-pz and fructose-1,6-p2 do not bind at the same competition is due to enzyme conformation differences, such that fructose-l,6-pz will not bind to the fructose-2,6-p2-binding conformation state and vice versa. Because phosphofructokinase from Ehrlich ascites tumor is very sensitive to activation by fructose-2,6-p2 but is unaffected by fructose-1,6-pz, Bosca et al. (20) suggested that isozymes sensitive to both bisphosphates have two different regulatory sites. On the other hand, although the yeast and erythrocyte phosphofructokinases are insensitive or less sensitive to activation by fructose-1,6-p2 than is the muscle enzyme, their activation by fructose-2,6-p2 can be antagonized Pettigrew and Frieden (19) recognized that their two-state model might be too limiting to fit all the facts; they themselves concluded that some of their experiments suggested that fructose-l,6-p2 stabilizes an enzyme conformation different from that stabilized by AMP.

4 7988 Activation of by fructose-1,6-p2, albeit at near-millimolar concentrations (21,22). Thus, the simplest explanation of the competition of the fructose bisphosphates is that they bind at the same site and that different isozymes have different degrees of specificity or differential affinity for the two isomers, with the muscle isozyme being among the least discriminating. This explanation is perhaps also supported by the linearity and common intersection point of double reciprocal plots for the binding of fructose-2,6-p2 to muscle phosphofructokinase in the presence of varying concentrations of fructose-1,6-p2 (13). Nevertheless, unequivocal resolution of this matter would require actual physical mapping of the respective binding sites, such as by x-ray crystallography. These kinetic experiments with fructose-1,6-pz and fructose-2,6-p2 indicate a subtle difference between binding and effectiveness of a ligand. Such a distinction has been noted previously with other ligands at other regulatory sites on phosphofructokinase. Thus, adenosine diphosphoribose and other dinucleotides can bind to the activating site for adenine mononucleotides and are as effective as the mononucleotide activators in protecting class I1 thiols; however, they are unable to activate the enzyme or to enhance the reactivity of the class I thiol group which is associated with the active state (23). In the case of bacterial phosphofructokinase, which is not sensitive to regulation by ATP, citrate, or AMP, x-ray crystallographic studies have demonstrated that the activator ADP and the inhibitor phosphoenolpyruvate both bind to the same regulatory site (24). Presumably the inhibition by phosphoenolpyruvate is due to a greater affinity for the less active conformation state of the enzyme, not just to competition with the activator, since the inhibition can be observed in the absence of ADP (25). The dissimilarity of action of fructose- l,6-pz and fructose-2,6-pz is an especially interesting case because the comparison is between two potent activators, rather than between an activator and an ineffective ligand or an inhibitor. If the fructose bisphosphates bind at the same site, then the biochemical mechanism for the observed kinetic differences must relate to the different orientation of a 1- phosphate versus a 2-phosphate with respect to the 6-phos- phate and the carbon chain or ring of the metabolite and therefore to the different spatial orientation and interaction with appropriate amino acid residues in the regulatory site. The relative positions of the highly charged phosphates are likely to be of particular importance. Although there might be a gross difference in binding of fructose-1,6-p2 and fructose-2,6-pz in terms of which chemical groups on the enzyme could be involved in the interactions with the activator and in transducing the binding into appropriate conformational change, even a seemingly small difference in orientation of the binding interactions might translate into a sufficient conformational difference to account for the observed kinetic effects. In this regard, the small movement of the heme iron involved in the change in oxygen affinity of hemoglobin is an instructive example (26). Again, elucidation of such details of mechanism would require physical mapping of enzyme-activator complexes. The difference between fructose-2,6-pz and fructose-1,6-pz as activators of phosphofructokinase has some analogy with previously reported differences between AMP and IMP as activators of phosphorylase b. For example, glucose-l-p af- fects the but not the Hill coefficient for AMP, but affects the Hill coefficient and not the for IMP (27). Binding studies indicated that binding of one molecule of AMP to phosphorylase b expels two molecules of the inhibitor glucose- 6-P, whereas binding of one molecule of IMP expels only one molecule of glucose-6-p (28). Such differences in kinetic and Phosphofructokinase by Fructose Bisphosphates physical properties with AMP versus IMP haveled these investigators to propose models of enzyme regulation for phosphorylase b with multiple conformation states. Phosphofructokinase would now seem to be a similar example. The physiological importance of activation of muscle phosphofructokinase by the fructose bisphosphates is uncertain at present. Epinephrine and insulin were reported to cause an increase in fructose-2,6-p2 in perfused hindlimb muscle and a stimulation of glycolysis; however, electrical stimulation increased glycolysis but decreased fructose-2,6-p2 levels and even abolished the increase in fructose-2,6-p2 induced by epinephrine (12). A transient increase in fructose-2,6-p* was observed in rat gastrocnemius muscle stimulated in situ at low frequencies, but not at higher frequencies, and there was no correlation with tissue lactate production (29). There was no change in fructose-2,6-p2 in similar experiments with extensor digitorum longus muscle. Foe et al. (11) pointed out that the concentration of fructose-2,6-p2 in muscle corresponds to only 8% of the concentration of phosphofructokinase subunits and therefore fructose-2,6-p2 would be of limited effectiveness even if fully bound to phosphofructokinase. Yet binding is likely to be incomplete, for under the conditions of our kinetic experiments, physiological concentrations of the inhibitors ATP (10 mm) and citrate (0.5 mm) raise the of muscle phosphofructokinase for fructose-2,6-p2 above 20 and 5 p~ in the presence of 10 and 100 pm AMP, respectively (data not shown), in other words,above the tissue concentration of 0.8 p~ fructose-2,6-pz measured by Hue et al. (12).2 On the other hand, even these apparently tiny effects may be of significance, since the glycolytic rate in muscle may be a small percentage of the maximum capacity of phosphofructokinase (14). Fructose-1,6-P2 content does increase with muscular activity. Much of the fructose-1,6-p2 is enzymebound, in particular to aldolase, which may obscure the magnitude of the rise in the free concentration (14). In cell-free extracts of skeletal muscle, activation of phosphofructokinase by fructose-1,6-p2 leads to oscillatory behavior of glycolysis, which may be of advantage in the maintenance of a high [ATP]/[ADP] ratio (3, 4). The oscillatory process involves the strong AMP dependence of the activation by fructose-l,6- Pz, such that high levels of fructose-1,6-p2 cannot compensate for lowered AMP concentrations. In this context, fructose- 2,6-P2 could not substitute for fructose-1,6-p,, because activation by the former does not have the autocatalytic nature of the latter and also because of the difference in kinetic interactions shown here, in particular with AMP. Whether fructose-2,6-p2 or fructose-1,6-pz or both are effective at activating phosphofructokinase in vivo, the actions of other regulatory metabolites such as citrate and AMP may be in part mediated through their strong effects on the saturation * Calculated on the basis that the intracellular water accounts for 50% of the wet weight and that the metabolites are evenly distributed throughout the intracellular water. On this basis, the skeletal muscle content of fructose-2,6-p2 reportedbykuwajima and Uyeda (30) would correspond to 5 PM. Their values for other tissues are 4-40 times greater than values given by Hue et al. (12). It is not clear whether these differences stem from unmentioned differences in animal strain or metabolic status or from the different assay methods used. Compared with results reported by others, the "normal" liver content of fructose-2,6-p2 reported by Kuwajima and Uyeda seems a bit high, whereas that reported by Hue et al. seems perhaps somewhat low. A glucose load only increased the liver content 70% for Kuwajima and Uyeda, but by over 400% for Hue et al. The bioluminescent assay of the metabolite used by the latter authors would seem to have fewer pitfalls than the older assay based on activation of phosphofructokinase, in view of possible interactive effects of other regulatory metabolites in the sample. Clearly there is a need for additional analyses of the fructose-2,6-p2 level in skeletal muscle under various conditions.

5 Activation of Phosphofructokinase by Fructose Bisphosphates 7989 curves for the two fructose bisphosphates. 13. Kitajima, S., and Uyeda, K. (1983) J. Biol. Chem. 258, Acknowledgment-The excellent technical assistance of Elizabeth 14. Tornheim, K., and Lowenstein, J. M. (1976) J. Biol. Chem. 251, Jensen is gratefully acknowledged. 15. Van Schaftingen, E., Jett, M.-F., Hue, L., and Hers, H.-G. (1981) Proc. Natl. Acad. Sci. U. S. A. 78, REFERENCES 16. Uyeda, K., Furuya, E., andluby, L. J. (1981) J. Biol. Chem. 256, 1. Bloxham, D. P., and Lardy, H. A. (1973) in The Enzymes (Boyer, P. D., ed) 3rd Ed., Vol. 8, pp , Academic Press, New 17. Kemp, R. G., and Foe, L. G. (1983) Mol. Cell. Biochem. 57, 147- York Uyeda, K. (1979) Adu. Enzymol. Relat. Area Mol. Biol. 48, Monod, J., Wyman, J., and Changeux, J. P. (1965) J. Mol. Biol , Tornheim, K., and Lowenstein, J. M. (1975) J. Biol. Chem. 250, 19. Pettigrew, D. W., and Frieden, C. (1979) J. Biol. Chem. 254, Boscti, L., Aragbn, J. J., and Sols, A. (1982) Biochem. Biophys. 4. Tornheim, K. (1979) J. Theor. Biol. 79, Res. Commun. 106, Hers, H.-G., and Van Schaftingen, E. (1982) Biochem. J. 206, 21. Bartrons, R., Van Schaftingen, E., Vissers, S., and Hers, H.-G (1982) FEBS Lett, 143, Uyeda, K., Furuya, E., Richards, C. S., and Yokoyama, M. (1982) 22. Heylen, A., Van Schaftingen, E., and Hers, H.-G. (1982) FEBS Mol. Cell. Biochem. 48, Lett Pilkis, S. J., El-Maghrabi, M. R., McGrane, M., Pilkis, J., Fox, 23. Gottscha1k;M.E., and Kemp, R. G. (1981)Biochemistry 20, E., and Claus, T. H. (1982) Mol. Cell. Endocr. 25, Hers, H. G., and Hue, L. (1983) Annu. Rev. Biochem. 52, Evans, p. R., and Hudson, p. J. (1979) Nature 279, Blangy, D., Buc, H., and Monod, J. (1968) J. Mol. Biol Claus, T. H., El-Maghrabi, M. R., Regen, D. M., Stewart, H. B., 35 McGrane, M., Kountz, P. D., Nyfeler, F., Pilkis, J., and Pilkis, 26. M. F. (lg7 ) S. J. (1984) Curr. Top. Cell. Regul. 23, Black, W. J., and Wang, J. H. (1968) J. Bwl. Chem. 243, Kemp, R. G. (1971) J. Biol. Chern. 246, Morange, M., Garcia Blanco, F., Vandenbunder, B., and Buc, H. 11. Foe, L. G., Lafshaw, S. P., and Kemp, R. G. (1983) Biochemistry (1976) Eur. J. Biochem. 66, , Minatogawa, Y., and Hue, L. (1984) Biochem. J Hue, L., Blackmore, p. F., Shikama, H., Robinson-Steiner, A.,30. Kuwajima, M., and Uyeda, K. (1982) Biochem. Biophys. Res. and Exton, J. H. (1982) J. Biol. Chem. 257, Commun. 104,84-88