Phosphofructokinase Depend on the Interaction Between

Home , Phosphofructokinase, Phosphotransferase

Proc. Natl. Acad. Sci. USA Vol. 91, pp. 5242-5246, June 1994 Biochemistry The cooperativity and allosteric inhibition of Escherichia coli phosphofructokinase depend on the interaction between threonine-125 and ATP (phosphoryl transfer/site-directed mutgenesis/enzyme regulation) ISABELLE AUZAT, GIStLE LE BRAS, AND JEAN-RENAUD GAREL* Laboratoire d'Enzymologie du Centre National de la Recherche Scientifique, 91198 Gif-sur-Yvette, France Communicated by Robert L. Baldwin, February 14, 1994

ABSTRACT During the reaction catalyzed by the phos- 9), comparing the crystal structures of the R and T states of phofructokinase (EC 2.7.1.11) from Escherichia coli, the phos- PFK does not suggest a structural mechanism for cooper- phoryl group transferred from ATP interacts with Thr-125 ativity. The three-dimensional structure ofunliganded E. coli [Shir a, Y. & Evans, P. R. (1988) J. Mol. Biol. 204, PFK resembles that of the R state (10), indicating that 973-994]. The replacement of Thr-125 by serine changes the cooperativity (as well as activation by ADP or GDP) could saturation by fructose 6-phosphate from cooperative to hyper- involve only this R state. Also, the PFKs from E. coli and bolic and abolishes the allosteric inhibition by phospho- Bacillus stearothermophilus have similar three-dimensional enolpyruvate. The same changes, a saturation by fructose structures, and both undergo a transition from the R to the T 6-phosphate that is no longer cooperative and an activity that state upon binding PEP (5, 6, 11), but the two enzymes have is no longer inhibited by phosphoenolpyruvate, are observed different saturations by Fru-6P: extremely cooperative forE. with wild-type phosphofructokinase when adenosine 5'-[r coli PFK (3) and hyperbolic for B. stearothermophilus PFK thio]triphosphate is used instead of ATP as the phosphoryl (12). These results indicate that the cooperativity of PFK donor. These two perturbations of the ATP-Thr-125 interac- does not seem coupled to the transition between the crys- tion lead to the suppression of both the allosteric inhibition by tallographic R and T states. phosphoenolpyruvate and the cooperativity of fructose-6- In the x-ray structure ofthe R state of PFK with substrates phosphate saturation, as ifreplacing the neutral oxygen ofATP and/or products bound, the phosphoryl group transferred by sulfur or removing the methyl group of Thr-125 had from ATP to Fru-6P interacts with Thr-125 (5, 7, 13). The "locked" phosphofructokinase in its active conformation. The hydroxyl group of Thr-125 is important since the Thr-125 geometry ofthis ATP-Thr-125 interaction and/or the presence Ala mutation decreases the activity of PFK by three orders of the methyl group on the P-carbon of Thr-125 are crucial for of magnitude and abolishes the cooperativity of Fru-6P the regulatory properties of phosphofructokinase. This inter- saturation (13). In this work, we have replaced either Thr-125 action could be a hydrogen bond between the neutral oxygen of by a Ser (Thr-125 -- Ser mutant) or ATP by adenosine the rphosphate of ATP and- the hydroxyl group of Thr-125. 5'-[ythio]triphosphate (ATP[(yS]). The main result is that these two perturbations result in the same changes in the The regulation of many biological processes involves allo- regulatory properties ofE. coli PFK: the saturation by Fru-6P steric interactions between distinct sites of the same protein, is no longer cooperative but is hyperbolic, and the protein no such that ligand binding at one site modifies the functional longer undergoes a transition into the inactive T state upon properties at a distant site. Cooperativity is observed when PEP binding. This suggests that the interaction between positive interactions take place between identical sites (1). Thr-125 and ATP is important for the allosteric behavior of One ofthe "classic" cooperative enzymes is Escherichia coli PFK. The Thr-125 -) Ser mutant of PFK still has a hydroxyl phosphofructokinase (PFK; ATP:D-fructose-6-phosphate group, but the absence of the methyl group on the (3-carbon 1-phosphotransferase, EC 2.7.1.11) (2, 3), which shows a probably affects the interaction with ATP. Similarly, the markedly sigmoidal saturation by its substrate fructose replacement of a single oxygen atom by sulfur in ATP[YS] is 6-phosphate (Fru-6P), with a Hill cooperativity coefficient a "mutation" of the yphosphate that perturbs its interaction close to nH = 4 for four Fru-6P sites (3). This enzyme is also with Thr-125. These results indicate that the side chain of sensitive to allosteric effectors, which bind to a regulatory Thr-125 and its interaction with the noncooperative substrate site remote from the active site; PFK is activated by ADP (or ATP are essential for the coupling between distant sites GDP) and inhibited by phosphoenolpyruvate (PEP) (3). responsible for cooperativity towards Fru-6P and allosteric The steady-state kinetics of E. coli PFK have been inter- inhibition by PEP of E. coli PFK. preted according to the concerted allosteric mechanism (4), in which the protein is in equilibrium between two states, an active R state, which binds Fru-6P and the activator ADP or MATERIALS AND METHODS GDP, and an inactive T state, which binds the inhibitor PEP The oligonucleotide 5'-GTCGATAGAGCCCGGGTA-3' (3). X-ray crystallography shows that PFK exists in two (Bioprobe Systems, Sous BQis, France) with a single mis- different conformations, R with activator and substrate (or match (underlined) was used to replace the ACT codon (Thr) products) bound (5) and T with inhibitor bound (6), which at position 125 by TCT (Ser). Site-directed mutagenesis was provides a structural explanation for the allosteric inhibition performed with the Amersham kit using Pvu I instead of Nci by PEP (6-8). However, in contrast with hemoglobin for contains an Nci I which the differences between the oxy and deoxy states I because the mutagenic oligonucleotide could largely explain the cooperativity of oxygen binding (8, Abbreviations: AMPPCP, adenosine 5'-[3,B,-methylene]triphosphate; ATP[(yS], adenosine 5'-[ythio]triphosphate; Fru-6P, D-fruc- The publication costs of this article were defrayed in part by page charge tose 6-phosphate; PEP, phosphoenolpyruvate; PFK, phosphofruc- payment. This article must therefore be hereby marked "advertisement" tokinase. in accordance with 18 U.S.C. §1734 solely to indicate this fact. *To whom reprint requests should be addressed. 5242 Downloaded by guest on September 25, 2021 Biochemistry: Auzat et A Proc. Natl. Acad. Sci. USA 91 (1994) 5243 site. The modified gene was completely sequenced to verify Table 2. Comparison of the enzymatic properties of wild-type the lack of any other alteration. Mutated and wild-type pikA PFK and the Thr-125 -. Ser mutant measured in both the absence genes inserted into the pDR540 tac vector (Pharmacia) were and presence of the allosteric activator GDP at 2 mM expressed after transforming cells of E. coli strain DF1020, Property Wild type Thr-125 Ser which are deleted for both pJk genes (14). Wild-type and mutant PFKs were purified as described (14, Without GDP using a coupled assay (16) in kat, sec-1 120 45 15). PFK activity was measured [Fru-6P]o.s, MM 340 140 a buffer composed of 0.1 M Mes, 0.051 M N-ethylmorpho- Km for ATP, p.M 57 17 line, and 0.051 M diethanolamine at pH 8.2. The ATP- nH 3.9 0.9 regenerating system composed of creatine phosphate and Inhibition by 10 82 <10 creatine kinase was omitted when ATP[yS] was used as mM PEP, % substrate. With 2 mM GDP The same buffer containing 10 mM Mg2e ion was used to kct, sec-1 140 30 measure the changes at 340 nm with excitation at 295 nm in Km for Fru-6P, p.M 41 170 intrinsic fluorescence of PFK upon binding Fru-6P, adeno- Km for ATP, p.M 60 21 sine 5'-[(3, -t-methylene]triphosphate (AMPPCP), or PEP as 0.9 reported (17). nH 1.0 [Fru-6P]o.s, half-saturating Fru-6P concentration. RESULTS AND DISCUSSION ATP (7, 13). The half-saturating Fru-6P concentration in the Fluorescence Measurements of Ligand Binding to the Thr- absence of GDP is not drastically altered by replacement of 125 -& Ser Mutant. Like that of wild-type PFK (17, 18), the Thr-125, with a 2.5-fold decrease for the Thr-125 -+ Ser fluorescence of the Thr-125 -+ Ser mutant changes upon the mutant (Table 2) and a 2.5-fold increase for the Thr-125 -- Ala binding of ligands. The fluorescence is increased by the mutant (13). binding ofAMPPCP (a nonhydrolyzable ATP analog) or PEP The prominent features of the Thr-125 -* Ser (Fig. 1 and and decreased by the binding of Fru-6P (Table 1). Fru-6P Table 2) and Thr-125 -+ Ala (13) mutants are their hyperbolic binding to wild-type PFK is not cooperative as seen from saturations by Fru-6P. The complete absence of cooperativ- equilibrium dialysis (17) and fluorescence (17, 18) measure- ity of the saturation by Fru-6P of the Thr-125 -- Ser mutant ments, and the fluorescence changes of the Thr-125 -> Ser is shown by the value of nH = 0.9 obtained for the cooper- mutant upon binding Fru-6P also show no cooperativity. The ativity coefficient using the Hill equation (Table 2). The mutation Thr-125 -> Ser does not drastically affect the methyl group on the (-carbon that distinguishes wild-type equilibrium constants for the binding of Fru-6P, AMPPCP, and the Thr-125 -) Ser mutant PFKs appears essential for and PEP (Table 1). The absence of cooperativity for Fru-6P cooperativity, even though it contributes only marginally to saturation or of PEP inhibition of the Thr-125 -* Ser mutant substrate binding and catalysis (Tables 1 and 2). However, described below are not due to major alterations in Fru-6P or Thr-125 is conserved in all bacterial PFKs, whether cooper- PEP binding. ative or not (19-23), showing that this methyl group is Steady-State Kinetic Properties of the Thr-125 -- Ser Mu- necessary but not sufficient for a cooperative behavior. Also, tant. Table 2 gives a comparison ofthe properties ofwild-type some eukaryotic PFKs have a Ser-125 in their active sites and PFK and the Thr-125 -) Ser mutant determined at steady still show cooperative Fru-6P saturation (24). state, in both absence and presence ofthe allosteric activator Regulator Properties of the Thr-125 -- Ser Mutant. Sur- GDP. The 2.5- to 3-fold reduction of the catalytic rate prisingly, the allosteric activator GDP does not increase the constant kct suggests that the Thr-125 -- Ser mutation causes affinity of the Thr-125 -* Ser mutant for Fru-6P but inhibits a slight perturbation ofthe OH group, whereas the Thr-125 -) Ala mutation, which removes this group, decreases kt by three orders of magnitude (13). Therefore the value of kcat depends on the presence and correct positioning of the hydroxyl group. 0.8 The side chain ofThr-125 does not significantly contribute to the affinity for ATP; for the Thr-125 -* Ala (13) and Thr-125 ;._ -* Ser (Table 2) mutants, Km for ATP is, respectively, 1.5 and 3 times lower than that for wild-type PFK. This result is 0.6 unexpected since x-ray crystallography suggests that the hydroxyl group of Thr-125 interacts with the -y-phosphate of ;- 9 0.4 Table 1. Dissociation constants and amplitudes of the fluorescence changes determined for the binding of Fru-6P, AMPPCP, and PEP to wild-type PFK (17) and the Thr-125 0.2 -* Ser mutant Ligand bound Wild type Thr-125 -* Ser Dissociation constant, p.M 0.0 0.3 0.5 0.8 1.0 1.3 1.5 Fru-6P 8 23 Fru-6P (mM) AMPPCP 5 9 PEP 500 1200 FIG. 1. Fru-6P saturation curves obtained for the Thr-125 -- Ser Fluorescence intensity* mutant with 1 mM ATP (e) and wild-type PFK with 1 mM ATP[yS] Fru-6P 80 93 (o) as substrate. In both cases the ordinate represents the fractional AMPPCP 110 115 saturation relative to maximum velocities, which are, respectively, 2.5 times (Table 2) and 400 times (Table 3) lower than that of PEP 108 110 wild-type PFK with 1 mM ATP. The dashed line shows the coop- *For both wild-type PFK and the Thr-125 -* Ser mutant, the erative saturation by Fru-6P of wild-type PFK measured with 1 mM fluorescence of the free protein was given an intensity of 100. ATP under the same conditions (nH = 3.9). Downloaded by guest on September 25, 2021 5244 Biochemistry: Auzat et al. Proc. Natl. Acad. Sci. USA 91 (1994) its activity (Table 2). The normal activation by GDP of Table 3. Comparison of the kinetic properties of wild-type PFK wild-type PFK involves a large increase in affinity for Fru-6P with ATP or ATP('yS] as a substrate (3) and a small increase in kat ofabout 1'5% (15). The increase Property ATP ATP[S] in kt caused by GDP can reach 50%6 oir 60% for some PFK mutants (unpublished results). The G]DP inhibition of the kat, sec-1 120 0.33 Thr-125 -* Ser mutant is not due to a cha,nge in Fru-6P affinity [Fru-6P]o.s, /AM 340 410 but to a decrease in k,,at ofabout 30% (TaLble 2). The influence Km for ATP, /AM 57 10 of GDP on knit suggests that GDP bin(ding induces a slight nH 3.9 0.8 change in the active site that improves the catalytic efficiency [Fru-6P]0.s, half-saturating Fru-6P concentration. of wild-type PFK and lessens that of the Thr-125 -* Ser mutant. The difference of 2.5- to 3-ft)ld in koat (Table 2) (Table 3), indicating that some of the enzyme-substrate suggests that the active sites of the mul iantand of wild-type interactions are slightly strengthened by the replacement of PFK do not have exactly the same conformation and could be one oxygen by sulfur. The half-saturating Fru-6P concentra- influenced in opposite ways by GDP bin(ding at the regulatory tion is not modified with as the phosphoryl donor site. (Table 3). However, the saturationATP[yS] by Fru-6P of wild-type Although it binds PEP with an affiniity similar to that of PFK is no longer sigmoidal but hyperbolic (Fig. 1). Cooper- wild-type PFK (Table 1), the Thr-125- Ser mutant is not ativity is not an intrinsic property of the enzyme; rather, it inhibited by PEP (Table 2 and Fig. 2): r teither the maximum involves the ternary complex between the enzyme and its two velocity nor the midpoint of its hyperbolic saturation by substrates and seems to be related to changes that are so Fru-6P are modified when up to 15 mM PEPois present. by subtle that they are affected by the replacement of one insensitivity to PEP inhibition of the TIPr-125is Ser mutant oxygen atom by sulfur on the -phosphate of ATP. It is is not due to a lack of PEP binding (see aboe.-Iaboveadand Tablemutable 1) tempting to attribute the disappearance of cooperativity with to the same cause as that with the Thr-125 -+ but probably to an inability to isomediize into the T state. ATP[yS]mutant (i.e., to an alteration ofthe ATP-Thr-125 interaction),Ser Without any detectable PEP inhibition and Fru-6P cooper- and the present results indicate that minor changes in this ativity, the Thr-125 -* Ser mutant behaLves as if it is unable interaction could shift the behavior of PFK from an extreme to undergo an allosteric transition into thie inactive T state and cooperative to a plain Michaelian saturation. is "locked" in the active R state, e)ven though its half- The overall cooperativity of E. coli PFK towards Fru-6P saturating Fru-6P concentration of 140 /M (Table 2) is 10 contains a significant kinetic contribution (7, 14, 15, 17) and times higher than that of 12.5 AM calcuLlated for the R state could thus be altered by a change in mechanism. The marked ofPFK (3). This mutant is the opposite ofthe other active site reduction in kcat suggests that the rate-limiting step might not mutants Ser-159 -- Asn and Thrl56 -e (GlySer) that are be the same when ATP[,yS] is used instead of ATP (Table 3). locked in the T state (25). The Thr-125 - Ser mutant seems Such a change in mechanism can hardly be invoked for the to be locked in its R state but shows the same fluorescence Thr-125 -* Ser mutant which is 35% to 40% as active as changes as wild-type PFK (Table 1), whiich indicates that the wild-type PFK (Table 2), and the hyperbolic Fru-6P satura- equilibrium between the low and high ftluorescence states is tion probably results from a structural and not from a kinetic not related to the allosteric transition between the R and T perturbation. states defined by x-ray crystallography The substitution ofATPby ATP[yS] as a substrate not only Steady-State Kinetic Properties of PFIK with ATP[(yS as a suppresses the cooperativity ofFru-6P saturation (Fig. 1) but Substrate. When ATP[yS] is used in:stead of ATP as a also abolishes PEP inhibition (Fig. 2). With ATP'yS] bound substrate, the catalytic rate constant kcaIt is reduced 400 fold in its active site, wild-type PFK becomes unable to undergo (Table 3). A similar decrease in kcat ha,s been measured for the allosteric transition into the inactive T state and remains yeast hexokinase (26), which suggests that the lower kcat locked in the active R state. It is not known whether this R reflects a lower reactivity of the thioptiosphate group. The state is unable to bind PEP or, like some mutants ofPFK, can affinity of PFK for ATP[ yS] is 5 times higher than for ATP bind PEP but without allosteric inhibition (27, 28). A Possible Hydrogen Bond Between the Hydroxyl Group of 120 Thr-125 and the Neutral Oxygen of the vPhosphate of AMP * q' Above neutral pH, the v-phosphate of ATP bears two negative charges, and the two negative oxygens probably inter- o act-with the two positive ligands, Mg2+ and Arg-72 (5, 7, 13). I-- The interaction with Thr-125 would thus involve the neutral oxygen, precisely that which is replaced by sulfur in ATP[yS], and a hydrogen bond between this neutral oxygen *a and the hydroxyl group of Thr-125 would be a plausible interaction This Q (5, 7). hydrogen bond remains to be char- acterized by x-ray crystallography for the Thr-125 -+ Ser Cla mutant, but if no hydrogen bond was present in this mutant, cc its kcat would be expected to decrease by 1000-fold, like that of the Thr-125 -> Ala mutant (13), and not by only 2.5-fold (Table 2). ~-- The v-Phosphate of ATP and the Allosteric Transition Between the R and T Structl States. The allosteric transi- 0.0 2.5 5.0 7.5 10.0 12.5 15.0 tion ofthe active site is suppressed by both the Thr-125--> Ser PEP (mM) mutation and the presence of ATP[yS], which suggests that the strength and/orgeometry ofthe Thr-125-ATP interaction FIG. 2. Inhibition by PEP of the Thr-125 ; Ser (e) and Arg-72 controls not only the cooperative behavior of PFK but also -- Ser (A) mutants measured at 2 mM Fru-6P and 1 mM ATP and of its ability to isomerize into the inactive T state. Among other wild-type PFK measured with 1 mM ATP[vySI (o) and 2 mM Fru-6P. changes, the transition from R into T involves breaking one The dashed line gives the PEP inhibition of wiId-type PFK measured electrostatic interaction between Arg-72 and the v-phosphate at 1 mM ATP and 2 mM Fru-6P under the siame conditions. of ATP and making one between Arg-72 and Glu-241 (Fig. 3) Downloaded by guest on September 25, 2021 Biochemistry: Auzat et al. Proc. Natl. Acad. Sci. USA 91 (1994) 5245

Rrg 72 Arg72- NH 2+ . .. OOC - Glu241

NH*2+

0l I ADP-0-P7% HO - Fru-6P 0* OH OH -Thrl 25 N+ I Thrl 25

FIG. 3. Partial representation of the changes occurring around the active site of PFK during the allosteric R -3 T transition. (Left) Ternary complex between the R state and the substrates ATP and Fru-6P, showing the interactions of the y-phosphate of ATP with Thr-125, Arg-72, and Mg2+ ion. (Right) T state with no ligand in the active site (PEP is bound to the regulatory site but is not shown), showing the ionic bond between Arg-72 and Glu-241 in the T state (inspired from refs. 6 and 7). (6-8). The relative stability of the R and T states depends (in Condusion. The side chain of Thr-125 plays a double key part) on the energy balance between the Arg-72-ATP and role in the properties of E. coli PFK: its hydroxyl group is Arg-72-Glu-241 interactions. That the Thr-125 -- Ser muta- critical for catalytic efficiency (13), and its methyl group is tion or the replacement of ATP by ATP[yS] lock PFK in the needed for cooperativity and allosteric inhibition. These two R state could be explained by a mutual influence between the major regulatory properties, cooperativity of Fru-6P satura- ligands ofthe y-phosphate ofATP. These perturbations ofthe tion and allosteric inhibition by PEP, depend crucially upon ATP-Thr-125 interaction could lead to a strengthening ofthe the strength and/or geometry of a specific enzyme-substrate Arg-72-ATP interaction, which would stabilize the R state interaction, possibly a hydrogen bond between the OH group and hinder or suppress the allosteric transition into the T ofThr-125 and the neutral oxygen ofthe 'y-phosphate ofATP. state. This ATP-Thr-125 interaction could be the first event that Substitutions of Glu-241 that weaken or suppress its inter- "triggers" PFK cooperativity through coupled movements action with Arg-72 are expected to destabilize the T state and of the hydroxyl group toward the -)ophosphate, ofthe methyl favor the R state. Indeed, the PFKs from Lactobacillus group and its neighboring residues, ofa subunit contact area, bulgaricus and Spiroplasma citri, which do not undergo an R etc., until the next active site, just as the oxygen-iron -- T transition (27, 29) have, respectively, Asp and Ile at interaction triggers hemoglobin cooperativity through cou- position 241 (22, 23) while Glu-241 is present in the PFKs pled movements of the iron into the heme plane, of the from E. coli (20), B. stearothermophilus (19), and Thermus proximal histidine, ofthe F-helix, ofthe FG corner, ofthe a(3 thermophilus (22), which respond to PEP binding by an R -- contact area, etc., until the next heme (2, 8, 9). The coop- T transition (3, 12, 30). It is more difficult to predict the effect erativity of PFK cannot yet be described by such a "path- of a mutation of Arg-72 since this residue is engaged in an way" of structural elements linking distant sites, but the ionic bond in both the R and T states, albeit with different present results suggest that only minor details will distinguish partners. Like all changes around the y-phosphate of ATP (7, a cooperative from a noncooperative structure or a flexible 13), the Arg-72 -- Ser mutation decreases the cooperativity protein able to oscillate between two structural states (R and of Fru-6P saturation (31). The Arg-72 -* Ser mutant is less T) from a rigid one locked in a unique state. inhibited by PEP than wild type (Fig. 2), indicating that the transition from R into T still takes place, but with a slight We are grateful to Dr. P. R. Evans for the kind gift of his Arg-72 stabilization of the R state. Ser mutant and for communication of his results prior to publi- Possible Kinetic Role of the Thr-125-ATP Interaction. The cation. This work has been supported by Grants ERS 029 from the Centre National de la Recherche Scientifique and 927-03 from the fact that the cooperativity ofPFK toward Fru-6P depends on Universit6 Paris 6. interactions with the y-phosphate of ATP, while ATP saturation is not cooperative (3), indicates that cooperativity 1. Monod, J., Changeux, J. P. & Jacob, F. (1%3) J. Mol. Biol. 6, pertains to aternary complex between the enzyme and its two 306-329. substrates. When measured at steady state, cooperativity is 2. Perutz, M. (1990) Mechanisms ofCooperativity and Allosteric related to the overall catalytic cycle, which involves binding Regulation in Proteins (Cambridge Univ. Press, Cambridge, of the two substrates, phosphoryl transfer, and release of U.K.). products. The reaction catalyzed by PFK follows a compli- 3. Blangy, D., Buc, H. & Monod, J. (1968) J. Mol. Biol. 31, 13-35. 4. Monod, J., Wyman, J. & Changeux, J. P. (1965) J. Mol. Biol. cated kinetic mechanism, and its cooperativity probably 12, 88-118. corresponds to a yet unknown combination of equilibrium 5. Shirakihara, Y. & Evans, P. R. (1988) J. Mol. Biol. 204, and rate constants (7, 14, 15, 17, 18). The crucial interaction 973-994. between Thr-125 and the transferred phosphoryl group that 6. Schirmer, T. & Evans, P. R. (1990) Nature (London) 343, controls cooperativity can be formed within the ternary 140-145. complex ofPFK with substrates, with products, and/or with 7. Evans, P. R. (1992) Proc. Robert A. Welch Found. Conf. the transition state and could even be present during the Chem. Res. 36, 39-54. 8. Perutz, M. (1989) Q. Rev. Biophys. 22, 139-236. entire catalytic cycle. The interaction with Thr-125 involves 9. Baldwin, J. M. & Chothia, C. (1979) J. Mol. Biol. 129, 183-191. not only the y-phosphate of ATP structurally but also the 10. Rypniewsky, W. R. & Evans, P. R. (1989) J. Mol. Biol. 207, phosphoryl acceptor kinetically, hence its role in the Fru-6P 805-821. dependence of the kinetics of the overall rate of phosphoryl 11. Evans, P. R., Farrants, G. W. & Hudson, P. J. (1981) Philos. transfer. Trans. R. Soc. London B 293, 53-62. Downloaded by guest on September 25, 2021 5246 Biochemistry: Auzat et al. Proc. Nadl. Acad. Sci. USA 91 (1994)

12. Valdez, B. V. C., French, B. A., Younathan, E. S. & Chang, 23. Branny, P., De La Torre, F. & Garel, J.-R. (1993) J. Bacteriol. S. H. (1989) J. Biol. Chem. 264, 131-135. 175, 5344-5349. 13. Berger, S. A. & Evans, P. R. (1992) Biochemistry 31, 9237-242. 24. Fothergill-Gilmore, L. A. & Michels, P. A. M. (1993) Prog. 14. Auzat, I. (1993) Ph.D. thesis (University of Paris, Paris). Biophys. Mol. Biol. 59, 105-235. 15. Deville-Bonne, D., Bourgain, F. & Garel, J.-R. (1991) Bio- 25. Kundrot, C. E. & Evans, P. R. (1991) Biochemistry 30, 1478- chemistry 30, 5754-5760. 1484. 16. Kotlarz, D. & Buc, H. (1982) Methods Enzymol. 90, 60-70. 26. Viola, R. E. & Cleland, W. W. (1978) Biochemistry 17, 4111- 17. Deville-Bonne, D. & Garel, J.-R. (1992) Biochemistry 31, 1695-1700. 4117. 18. Berger, S. A. & Evans, P. R. (1991) Biochemistry 30, 8477- 27. Auzat, I., Le Bras, G., Branny, P., De La Torre, F., Theunis- 8480. sen, B. & Garel, J.-R. (1994) J. Mol. Biol., 235, 68-72. 19. Kolb, E., Hudson, P. J. & Harris, J. I. (1980) Eur. J. Biochem. 28. Serre, M. C., Teschner, W. & Garel, J.-R. (1990)J. Biol. Chem. 108, 587-597. 265, 12146-12148. 20. Hellinga, H. W. & Evans, P. R. (1985) Eur. J. Biochem. 149, 29. Le Bras, G., Deville-Bonne, D. & Garel, J.-R. (1991) Eur. J. 363-373. Biochem. 198, 683-687. 21. Xu, J., Seki, M., Denda, K. & Yoshida, M. (1991) Biochem. 30. Xu, J., Oshima, T. & Yoshida, M. (1990) J. Mol. Biol. 215, Biophys. Res. Commun. 176, 1313-1318. 597-606. 22. Chevalier, C., Saillard, C. & Bovd, J. (1990) J. Bacteriol. 172, 31. Berger, S. A. & Evans, P. R. (1990) Nature (London) 343, 3693-3703. 575-576. Downloaded by guest on September 25, 2021