Studies of the allosteric properties of maize leaf phosphoenolpyruvate carboxylase with the phosphoenolpyruvate analog phosphomycin as activator

Studies of the allosteric properties of maize leaf phosphoenolpyruvate carboxylase with the phosphoenolpyruvate analog phosphomycin as activator

Biochimica et Biophysica Acta 1386 (1998) 132^144 Studies of the allosteric properties of maize leaf phosphoenolpyruvate carboxylase with the phospho...

328KB Sizes 0 Downloads 34 Views

Biochimica et Biophysica Acta 1386 (1998) 132^144

Studies of the allosteric properties of maize leaf phosphoenolpyruvate carboxylase with the phosphoenolpyruvate analog phosphomycin as activator Carlos Mu¨jica-Jime¨nez, A. Castellanos-Mart|¨nez, R.A. Mun¬oz-Clares * Departamento de Bioqu|¨mica, Facultad de Qu|¨mica, Universidad Nacional Auto¨noma de Me¨xico, Mexico, DF 04510, Mexico Received 10 February 1998; accepted 23 April 1998

Abstract The antibiotic phosphomycin (1,2-epoxypropylphosphonic acid), an analog of phosphoenolpyruvate (PEP), behaved not as an inhibitor, but as an activator, of the enzyme phosphoenolpyruvate carboxylase (PEPC) from maize leaves. Multiple activation studies indicated that the analog binds to the Glc6P-allosteric site producing a more activated enzyme than Glc6P itself. Because of this, we used phosphomycin as a tool to further extend our understanding of the mechanisms of allosteric regulation of C4 -PEPC. Initial velocity data from detailed kinetic studies, in which the concentrations of free and Mgcomplexed PEP and phosphomycin were controlled, are consistent with: (1) the true activator is free phosphomycin, which competes with free PEP for the Glc6P-allosteric site ; and (2) the Mg-phosphomycin complex caused inhibition by binding to the active site in competition with MgPEP. Therefore, although the Glc6P-allosteric site and the active site are able to bind the same ligands, they differ in the form of substrate and activator they bind. This important difference allows the full expression of the potential of activation and prevents inhibition by the activators, including the physiological ones, which are mostly uncomplexed at physiological free Mg2‡ concentrations. At fixed low substrate concentrations, the saturation kinetics of the enzyme by phosphomycin showed positive cooperativity at pH 7.3 and 8.3, although at the latter pH, the kinetics of saturation by the substrate was hyperbolic. The cosolute glycerol greatly increased the affinity of the enzyme for phosphomycin and abolished the cooperativity in its binding, but did not eliminate the heterotropic effects of the activator. Therefore, the heterotropic and homotropic effects of the activator are not always coupled to the homotropic effects of the substrate, which argues against the two-state model previously proposed to explain the allosteric properties of maize-leaf PEPC. ß 1998 Elsevier Science B.V. All rights reserved. Keywords: Phosphoenolpyruvate carboxylase; Activation; Phosphomycin; Glucose-6-phosphate allosteric site; Glycine ; Glycerol ; (Zea mays L.)

1. Introduction

Abbreviations: Glc6P, glucose-6-phosphate; fMg2‡ , free Mg2‡ ; PEP, phosphoenolpyruvate; fPEP, free phosphoenolpyruvate; PEPC, phosphoenolpyruvate carboxylase * Corresponding author. Fax: +52 (5) 622-5329; E-mail: [email protected]

Phosphoenolpyruvate carboxylase (PEPC, EC 4.1.1.31) is recognized as a key enzyme in the CO2 assimilation pathway of C4 and CAM plants. PEPC from the C4 plant Zea mays is a regulatory enzyme subjected to interacting covalent [1,2] and allosteric control. Regarding the latter, its activity is increased

0167-4838 / 98 / $19.00 ß 1998 Elsevier Science B.V. All rights reserved. PII: S 0 1 6 7 - 4 8 3 8 ( 9 8 ) 0 0 0 9 3 - 4

BBAPRO 35674 14-7-98

C. Mu¨jica-Jime¨nez et al. / Biochimica et Biophysica Acta 1386 (1998) 132^144

by glucose-6-phosphate (Glc6P) [3,4] and glycine [4^ 6], and decreased by L-malate [7]. These three compounds likely have separate binding sites on the enzyme [8,9]. Besides Glc6P, other phosphate esters, such as fructose-6-phosphate, fructose-2,6-bisphosphate, dihydroxyacetone-phosphate [10], 3-phosphoglycerate [11], glycerol phosphate [9], AMP [12], and acetyl phosphate [13], also behave as C4 -PEPC activators. None of these compounds show clear structural similarity to the substrate phosphoenolpyruvate (PEP) other than having a phosphate group. However, some PEP analogs, such as methyl 2-dihydroxyphosphinoylmethyl-2-propenoate [14], sulfoenolpyruvate [15], and phenylphosphate [16], are known to be activators of PEPC, while other PEP analogs behave as competitive inhibitors of this enzyme ([17] and references therein, [18]). The distinction between activation and inhibition is not clearly de¢ned since certain phosphorylated compounds, as carbamyl phosphate [13] and 3-phosphoglycerate [10,11], may behave as activators or inhibitors. Even more, PEP itself binds to the Glc6P allosteric site, thus behaving both as substrate and allosteric activator of the PEPC-catalyzed reaction [16,19,20]. The molecular basis for the di¡erent behavior of the PEP analogs is not known yet, although it is clear that such knowledge may shed light on the mechanisms of catalysis and regulation of this enzyme. We have now studied the e¡ect on the activity of maize leaf PEPC of another PEP analog, 1,2-epoxypropylphosphonic acid (phosphomycin [21]) whose structure is compared to that of PEP in Scheme 1. Phosphomycin is widely used as an antibacterial agent [22] due to its ability to irreversibly inhibit the enzyme UDP-N-acetylglucosamine 1-carboxyvinyltransferase (EC 2.5.1.7) [23,24], which catalyzes the ¢rst committed step in the biosynthesis of the

133

nucleotide muramyl peptides which serve as cellwall precursors in bacteria. We report here that phosphomycin acts as a potent, reversible non-essential activator of PEPC from maize leaves by binding to the same allosteric site as Glc6P and other PEP analogs which behave as activators. In view of these results, we used phosphomycin as a tool to further extend our understanding of the mechanisms of allosteric regulation of C4 -PEPC. The kinetics of activation of PEPC by Glc6P, or by any of the other ligands of the Glc6P site, have been studied so far using total PEP as the variable substrate at ¢xed, high total concentrations of the metal ion, either Mg2‡ or Mn2‡ . Although it is has been reported that PEP binds to the active site of maize leaf PEPC after Mg2‡ is bound [25], several other reports show that the preferential substrate of this enzyme is MgPEP [16,20,26,27]. Given the stability constant of the MgPEP complex [28^30], the intracellular levels of MgPEP will be less than 10% of the total PEP at 0.4 mM free Mg2‡ (fMg2‡ ), which is the estimated concentration of fMg2‡ in the cytoplasm of plant cells [31]. Similarly, the physiological activators will also be uncomplexed under the intracellular conditions since their stability constants are even lower than that of PEP [32]. However, it has been proposed that the MgGlc6P complex is the true activator of the enzyme [33]. Because of that, we considered it of interest to carry out detailed kinetic experiments using MgPEP as the variable substrate and controlling variables such as the concentration of free PEP (fPEP), fMg2‡ , or of free or complexed activator, which were not taken into account in previous kinetic studies, in order to investigate further the properties of the allosteric Glc6P site. 2. Materials and methods 2.1. Chemicals and biochemicals

Scheme 1. Comparison of the structures of phosphomycin and PEP.

PEP (monocyclohexylammonium salt), NADH (disodium salt), porcine heart malic dehydrogenase, and phosphomycin were purchased from Sigma (St. Louis, MO). Glycerol and EDTA (disodium salt) were from Merck (Darmstadt, Germany). All other chemicals of analytical grade were from standard suppliers.

BBAPRO 35674 14-7-98

134

C. Mu¨jica-Jime¨nez et al. / Biochimica et Biophysica Acta 1386 (1998) 132^144

2.2. Phosphoenolpyruvate carboxylase extraction and puri¢cation The enzyme was extracted from leaves of maize (Zea mays L., cv. Chalquen¬o) and puri¢ed as described elsewhere [34]. The ¢nal speci¢c activity of the enzyme preparation used was 33 Wmol/min/mg protein (assayed at 30³C, pH 7.3, 5 mM total PEP, and 10 mM total Mg2‡ ). The I50 for malate of this enzyme preparation, determined as described by Wang et al. [35], was 0.14 mM, indicative of a non-phosphorylated, non-truncated protein. This value did not decrease after exhaustive incubation with alkaline phosphatase, con¢rming the non-phosphorylated state of the enzyme [36]. Incubation with protein kinase A, following the procedure described by Du¡ et al. [36], increases the I50 for malate to 2.2 mM, indicative of the integrity of the NH2 -terminal region of the protein.

maintained to within 30 þ 0.1³C with a circulating water bath and thermospacers. Assay points were performed at least in duplicate and averaged. One unit of activity corresponds to 1 Wmol of NADH produced per min. Protein concentration was determined by the method of Bradford [37] using bovine serum albumin as standard. 2.4. Data analysis PEPC kinetic data were analyzed by non-linear regression calculations using commercial computing programs formulated with the algorithm of Marquardt [38]. Kinetic data depending upon varied concentration of substrate were ¢tted to the Michaelis^ Menten equation (Eq. 1) for hyperbolic kinetics, to the Hill equation (Eq. 2) for sigmoidal kinetic, or to the substrate inhibition equation (Eq. 3), v ˆ V max US=…K m ‡ S†;

…1†

2.3. Enzyme assay

v ˆ V max USn =…S n0:5 ‡ Sn †;

…2†

PEP carboxylase activity was measured spectrophotometrically at 30³C with a coupled enzyme assay using malate dehydrogenase, and following the oxidation of NADH at 340 nm with a Beckman DU7500 spectrophotometer equipped with a kinetics software package. The standard reaction mixture contained, in a ¢nal volume of 0.5 ml, 100 mM triethanolamine-HCl bu¡er (pH 7.3 or 8.3), 1 mM EDTA, 10 mM NaHCO3 , 0.2 mM NADH, 2 units of malate dehydrogenase, and the amounts of total Mg2‡ (as MgCl2 ), total PEP and total phosphomycin needed to give the desired concentrations of MgPEP, Mg-phosphomycin and of any of the free species [16]. The Mg-phosphomycin stability constant used (0.237 mM31 ) was determined by us under the conditions of the enzymatic assay by competition with 8hydroxyquinoline, as described in [32]. We use a stability constant of the MgPEP complex of 0.180 mM31 [28], since it was very close to that determined by us under our conditions of assay. Other stability constant values used were those described in [16]. The pH of the PEP solution was previously adjusted to the pH of the assay using appropriate amounts of triethanolamine. The reaction was started by addition of the enzyme. Assays were conducted in 1.0 cm path-length cuvettes and the temperature was

v ˆ V max US=…K m ‡ S ‡ S 2 =K i †;

…3†

where v is the experimentally determined initial velocity, Vmax the maximum velocity, S the concentration of the variable substrate, Km the Michaelis^ Menten constant for the substrate S, S0:5 the concentration of substrate that gives half-maximum velocity, Ki the inhibition constant for the substrate and n the Hill number. In the experiments in which the concentration of the activator was varied at constant concentration of substrates, Eqs. 4^6 were used to ¢t the data to hyperbolic or sigmoidal saturation curves. When deactivation was observed Eqs. 7^9 were used. v ˆ …vamax 3v0 †A=…A0:5 ‡ A† ‡ v0 ;

…4†

…va 3v0 †=v0 ˆ Actmax UA=…A0:5 ‡ A†;

…5†

…va 3v0 †=v0 ˆ Actmax UAn =…An0:5 ‡ An †;

…6†

v ˆ …vamax 3v0 †A=…A0:5 ‡ A ‡ A2 =K i † ‡ v0 ;

…7†

v ˆ …vamax 3v0 †An =…An0:5 ‡ An ‡ A…n‡1† =K i † ‡ v0 ;

…8†

…va 3v0 †=v0 ˆ Actmax UA=…A0:5 ‡ A ‡ A2 =K i †;

…9†

where v0 and va are the initial velocities in the ab-

BBAPRO 35674 14-7-98

C. Mu¨jica-Jime¨nez et al. / Biochimica et Biophysica Acta 1386 (1998) 132^144

sence and presence of activator, respectively, vamax is the highest velocity obtained at saturating activator concentrations, Actmax is the maximum activation, i.e. the maximum increase in enzyme activity measured as times the non-activated activity, A is the activator concentration, and A0:5 is the concentration of activator that gives half-maximum activation at a ¢xed concentration of substrate (it is therefore an apparent dissociation constant for the activator depending on [S]/Km by a factor which varies with the activation mechanism). At pH 8.3, the activation constant, Ka , and the limiting Vmax and Km values were determined by Eqs. 10 and 11, respectively [39], appV max ˆ V max …1 ‡ LA=KK a †=…1 ‡ A=KK a †

…10†

appK m ˆ K m …1 ‡ A=K a †=…1 ‡ A=KK a †;

…11†

where L and K are interaction factors that describe the in£uence that the binding of the activator has on the catalytic step or on the binding of the substrate, respectively, LVmax and KKm are the maximum velocity and the Michaelis^Menten constant at saturating concentration of activator, and KKa the limiting activation constant at saturating concentration of the substrate.

3. Results 3.1. Multiple activation studies The antimicrobial e¡ect of phosphomycin is based on its ability to bind to the active site of the enzyme UDP-N-acetylglucosamine 1-carboxyvinil transferase, as an analog of the substrate PEP, and covalently modify a Cys residue [23,24]. Since Cys residues had been reported to be at the active site of maize leaf PEPC [40,41], we were originally interested in testing this compound as a possible a¤nity label for mapping the active site. However, phosphomycin did not behave as an inhibitor of PEPC, either reversible or irreversible, but rather as an activator. As shown in Fig. 1A, 5 mM phosphomycin was a more potent activator than 5 mM Glc6P or 5 mM of other phosphorylated compounds which have been previously shown to bind to the Glc6P allosteric site, such as acetyl phosphate, phenylphosphate,

135

fructose-6-phosphate, myo-inositol phosphate, fructose-1,6-bisphosphate, and fructose-2,6-bisphosphate. The degree of PEPC activation by phosphomycin was decreased when assayed together with a less potent activator (Fig. 1A). Some of them, as the bisphosphate esters of fructose, were very poor activators but very e¤cient in decreasing the activation induced by phosphomycin. These results suggest that the activators bind to the same site and that the di¡erences between phosphomycin and the other activators could be mainly due to a higher potency of phosphomycin as activator once bound to the enzyme, rather than to a higher degree of saturation of the enzyme by phosphomycin. To investigate this point, we determined the saturation of the enzyme by phosphomycin and by the physiological activator Glc6P at pH 7.3, saturating HCO3 3 (10 mM) and non-saturating Mg2‡ and PEP (total concentrations of 0.5 and 1.5 mM, respectively). We found that the a¤nity of the enzyme for phosphomycin is almost identical to that for Glc6P (A0:5 values 2.8 þ 0.3 and 2.7 þ 0.3 mM, respectively), but the former activator produced a much more active enzyme than the latter (the maximum activation values, estimated by ¢tting the data in Fig. 1B to Eq. 8, for phosphomycin and for Glc6P were 33.3 þ 2.4 and 12.5 þ 1.0 times the control activities, respectively). When the response of the enzyme to Glc6P in the presence of 1 mM phosphomycin was determined, we observed additive e¡ects of both activators at low concentrations of Glc6P, but as the concentration of Glc6P further increased, the enzyme was deactivated until reaching the activation level of the Glc6P-saturated enzyme (Fig. 1B), what suggests a common binding site for both activators. In independent experiments, in which we measured the protection a¡orded by ligands of the enzyme against the desensitization to activation by Glc6P caused by pyridoxal-5P-phosphate [34], we found that 10 mM phosphomycin afforded total protection, while 10 mM Glc6P a¡orded only around 60% protection (Tovar-Me¨ndez and Mun¬oz-Clares, unpublished results). This experimental evidence provides additional support to our conclusions that phosphomycin binds to the Glc6P allosteric site and that there might be di¡erences between Glc6P and phosphomycin in the putative conformational change that each activator elicited upon binding to the enzyme.

BBAPRO 35674 14-7-98

136

C. Mu¨jica-Jime¨nez et al. / Biochimica et Biophysica Acta 1386 (1998) 132^144

Fig. 1. Multiple activation studies of maize leaf PEPC. (A) Relative activation by 5 mM of the following compounds: phosphomycin (Pmyc), Glc6P, acetyl phosphate (AcP), phenylphosphate (xP), fructose-6-phosphate (F6P), myo-inositol phosphate (InsP), fructose1,6-bisphosphate (F16bP), and fructose-2,6-bisphosphate (F26bP), in the absence (white bars) and presence (black bars) of 5 mM phosphomycin. The puri¢ed enzyme was assayed at 30³C in 100 mM triethanolamine-HCl bu¡er, pH 7.3, containing 3.27 mM total PEP and 1.5 mM total Mg2‡ . (B) Activation by Glc6P (a), phosphomycin (R), and Glc6P plus 1 mM phosphomycin (b). Enzyme activity was measured as in (A), but at 1.5 mM total PEP and 0.5 mM total Mg2‡ . (C) Activation by glycine in the absence (E) and presence (F) of 1 mM phosphomycin. Assays were carried out as in (B). The initial velocity data in (B) and (C) were ¢tted to Eq. 7 (b,F), Eq. 8 (a,R) or Eq. 4 (E). The activity of the controls (v0 , no activator added) was 8.1 units/mg protein for (A) and 0.74 units/ mg protein for (B) and (C).

Phosphomycin is also a more potent activator of maize leaf PEPC than glycine, since the maximum activation elicited by the latter was only of 14.8 þ 0.4 times the control activity, as estimated by a ¢t of the data in Fig. 1C to Eq. 4. But contrary to that observed with Glc6P, we found a summation of the e¡ects of glycine and phosphomycin when we looked at the response of the enzyme to glycine plus 1 mM phosphomycin (Fig. 1C), clearly indicating that both activators bind to di¡erent allosteric sites. This concentration of phosphomycin greatly reduces the A0:5 for glycine, from 4.1 þ 0.6 mM in its absence to 0.6 þ 0.1 mM in its presence, and also the A0:5 for Glc6P, from 2.7 þ 0.3 mM to 0.3 þ 0.1 mM. The former result is in accordance with the reported ability of glycine to induce an in-

creased a¤nity of the enzyme for Glc6P [42], but the latter result may seem surprising since competitive binding of phosphomycin and Glc6P to the same allosteric site should have opposite e¡ects. However, taking into account that phosphomycin greatly decreases the Km for the substrate (as it will be shown below), and that the A0:5 values estimated from a ¢t of the data to Eqs. 4^9 are, in fact, apparent dissociation constants which decrease as the [S]/Km ratio increases, the e¡ects of phosphomycin on the apparent a¤nity of the other activators are most likely due to the increased saturation of the enzyme by the substrate in the presence of phosphomycin. Therefore, binding experiments in the absence of substrate are required in order to know whether ligands of the Glc6P allosteric site directly increase the a¤nity of

BBAPRO 35674 14-7-98

C. Mu¨jica-Jime¨nez et al. / Biochimica et Biophysica Acta 1386 (1998) 132^144

137

the glycine allosteric site for its ligands and vice versa. Under our experimental conditions, phosphomycin neither produced any change in the activation state of the enzyme after prolonged incubation nor caused its desensitization to Glc6P (not shown). Thus, it seems that the Glc6P-allosteric site does not have any reactive amino acid residue able to covalently react with this compound, although Cys residues appear to be involved in the response of maize leaf PEPC to Glc6P [43]. 3.2. E¡ect of phosphomycin on the kinetics of saturation of phosphoenolpyruvate carboxylase by MgPEP In order to characterize the PEPC activation by phosphomycin more completely and to extend further our understanding of the mechanism of allosteric regulation of the enzyme, we studied the e¡ect of several phosphomycin concentrations (0.25. 0.5 and 1 mM) on the saturation kinetics of maize leaf PEPC by MgPEP at saturating concentrations of HCO3 3 (10 mM) at pH 7.3. MgPEP was varied in a range from 0.1 to 4 mM, at two ¢xed concentrations of fPEP (5 and 0.5 mM). The initial velocity patterns (Fig. 2) showed that phosphomycin has qualitatively similar e¡ects regardless of the concentration of fPEP; it greatly decreased appKm (MgPEP) and increased appVmax to a much lesser extent. However, quantitatively important di¡erences were found between the kinetic parameters obtained at the two ¢xed fPEP. Thus, the e¡ects of 0.25 mM phosphomycin were much greater at 0.5 mM than at 5 mM fPEP (Fig. 2, open circles), suggesting competitive binding of phosphomycin and fPEP to the allosteric site. In the ¢rst case (Fig. 2B), 0.25 mM phosphomycin increased the appVmax /Km (MgPEP) around 4times, from 61.9 to 257.3 units mg protein31 mM31 , and abolished the positive homotropic cooperativity observed in its absence, while in the second (Fig. 2A) 0.25 mM phosphomycin only increased the appVmax / Km (MgPEP) around 2-times from 81.7 to 178.1 units mg protein31 mM31 , and the binding of substrate was still cooperative (n = 1.35 þ 0.08) in the presence of the activator. Moreover, while at 5 mM PEP (Fig. 2A) increasing the phosphomycin concentration above 0.25 mM lead to progressive increases in

Fig. 2. Kinetics of saturation by MgPEP of maize-leaf PEPC in the absence (b) and presence of 0.25 (a), 0.5 (R), and 1 (F) mM phosphomycin. Assays were carried out at pH 7.3, keeping constant the concentration of fPEP at 5 mM (A) or 0.5 mM (B) and varying the concentrations of MgPEP as indicated (fMg2‡ was varied from 0.11 to 4.44 mM in (A) and from 1.11 to 44.4 mM in (B)). Other assay conditions were as in Fig. 1. The points are the experimentally determined values, and the lines drawn through these points are calculated from the best ¢t of the data to Eq. 1, Eq. 2 or Eq. 3.

appVmax /Km (MgPEP) as a result of increased appVmax and reduced appKm (MgPEP), at 0.5 mM fPEP (Fig. 2B), concentrations of phosphomycin higher than 0.25 mM not only further reduced the appKm (MgPEP) values but also the appVmax values. The anomalous behavior of phosphomycin under these conditions in which fMg2‡ concentrations are high (ranging from 1.11 to 44.4 mM) is probably the consequence of the presence of kinetically signi¢cant levels of the Mg-phosphomycin complex, which may be a competitive inhibitor of the substrate MgPEP. If this is the case, the appVmax and appKm (MgPEP)

BBAPRO 35674 14-7-98

C. Mu¨jica-Jime¨nez et al. / Biochimica et Biophysica Acta 1386 (1998) 132^144

138

Fig. 3. Double-reciprocals plots of the kinetics of saturation by MgPEP of maize-leaf PEPC in the absence (b) or presence of 0.25 (a), 0.4 (R), 1 (F), and 5 mM (E) phosphomycin. Assays were carried out at pH 8.3, keeping constant the concentration of fMg2‡ at 1 mM and varying the concentration of MgPEP as indicated (fPEP was varied from 0.13 to 22.2 mM). Other assay conditions were as in Fig. 1. Inset: Replots of appVmax (b) and appKm (MgPEP) (a) versus phosphomycin concentration. The lines are the best ¢t to Eq. 10 or Eq. 11, respectively.

values should be reduced by the same factor as the concentration of ¢xed, total phosphomycin is raised, since the concentrations of the putative competitive inhibitor, Mg-phosphomycin, increased simultaneously and in a constant ratio to those of the substrate, MgPEP [44]. However, our results show a

larger reduction in appKm (MgPEP) than in appVmax when the ¢xed total phosphomycin was increased, which lead to marked increases in appVmax / Km (MgPEP). This discrepancy with the expected results may be accounted for by the summation of the e¡ects of phosphomycin under these conditions: the

Table 1 E¡ect of MgPEP, fPEP and fMg2‡ on the saturation kinetics of maize leaf PEPC by phosphomycin Kinetic parametersa

Assay conditions [MgPEP] (mM) pH 7.3 0.1 0.1 2.0 2.0 pH 8.3 0.1 0.1 2.0 2.0 a

[fPEP] (mM)

[fMg2‡ ] (mM)

v0 (units/mg protein)

Maximum activation (times)

A0:5 (WM)

n

0.5 5.0 0.5 5.0

1.11 0.11 22.2 2.22

1.3 1.3 18.2 17.4

16.2 þ 0.9 15.3 þ 1.6 0.67 þ 0.02 0.85 þ 0.08

393 þ 56 1340 þ 296 21 þ 3 483 þ 12

1.43 þ 0.26 1.28 þ 0.18 1 1

0.5 5.0 0.5 5.0

1.11 0.11 22.2 2.22

2.8 2.8 14.1 15.2

5.9 þ 0.1 5.7 þ 0.2 0.52 þ 0.01 0.50 þ 0.02

233 þ 16 857 þ 62 21 þ 7 283 þ 41

1.69 þ 0.20 1.68 þ 0.14 1 1

Values þ S.E. were estimated by the best ¢t to Eq. 5 or Eq. 6.

BBAPRO 35674 14-7-98

C. Mu¨jica-Jime¨nez et al. / Biochimica et Biophysica Acta 1386 (1998) 132^144

large reduction in appKm (MgPEP) and small increase in appVmax brought about by binding of free phosphomycin to the allosteric site, and the equivalent decreases in appKm (MgPEP) and appVmax produced by binding of Mg-phosphomycin to the active site. At pH 8.3, phosphomycin produced similar e¡ects on the kinetics of saturation of the enzyme by MgPEP to those at pH 7.3, in spite that at this pH there are no homotropic e¡ects of the substrate, i.e. saturation by MgPEP follows Michaelian kinetics. Fig. 3 shows the linear double reciprocal plots of 1/ v versus 1/[MgPEP] obtained by varying MgPEP from 0.025 to 4 mM at ¢xed 1 mM fMg2‡ and at several ¢xed concentrations of phosphomycin (from 0.25 to 5 mM, total concentrations). This pattern is indicative of mixed non-essential activation, and in fact we obtained a good ¢t of the replots of appVmax and appKm from Fig. 3 versus [phosphomycin] to Eqs. 10 and 11, yielding estimated values of 24.5 þ 0.5 units/mg protein for Vmax , 0.20 þ 0.02 mM for Km (MgPEP), 3.32 þ 0.60 mM for the activation constant, Ka , 0.08 þ 0.01 for K, and 1.46 þ 0.04 for L. Using these estimates, we calculated a limiting Ka at saturating substrate concentration, KKa; of 0.25 mM, and limiting values of Vmax , LVmax , and Km (MgPEP), KKm (MgPEP), of 35.8 units/mg protein and of 0.016 mM, respectively, both at in¢nite activator concentration. The estimated value for the interaction factor K suggests that, under these experimental conditions, there is a 13-fold increase in the binding of the activator to the allosteric site when the active site has been already occupied, and vice versa. 3.3. E¡ect of MgPEP, fPEP and fMg2+ on the kinetics of saturation of phosphoenolpyruvate carboxylase by phosphomycin The above results suggested a competition between fPEP and phosphomycin for binding to the activator site. To test this possibility, the saturation of maize leaf PEPC by phosphomycin (from 0 to 5 mM, total concentrations) was studied at subsaturating (0.1 mM) and almost saturating (2 mM) MgPEP, at pH 7.3 and 8.3. In these experiments, the fPEP concentrations were 0.5 and 5 mM, and therefore fMg2‡ concentrations varied in a range from 0.11 to 22.2

139

Fig. 4. E¡ect of free phosphomycin and Mg-phosphomycin on the activity of maize-leaf PEPC. Assays were carried out at pH 7.3, keeping constant the concentration of MgPEP at 0.1 mM and varying the concentrations of free phosphomycin and Mgphosphomycin as indicated in the ¢gure. Accordingly, fPEP was raised from 0.1 to 0.5 mM and fMg2‡ decreased from 5.55 to 1.11 mM when free phosphomycin was raised, and, vice versa, fPEP was decreased from 0.1 to 0.02 mM and fMg2‡ increased from 5.55 to 27.75 mM, when Mg-phosphomycin was raised. Other assay conditions were as in Fig. 1. Numbers in parentheses below each bar indicate the enzyme activity (as units/mg protein) in the absence of phosphomycin.

mM. The results obtained at both pH values were very similar and are summarized in Table 1. The maximum activation achieved by phosphomycin was dependent on the MgPEP concentration, being similar at the same [MgPEP], regardless of the concentration of the free species, and much greater at low than at high [MgPEP]. This ¢nding is consistent with the e¡ects of phosphomycin being much greater on Km (MgPEP) than on Vmax . However, the A0:5 showed a clear dependency not only on the MgPEP concentration but also on the concentrations of the free species. Thus, the A0:5 greatly decreased as the concentration of MgPEP increases, in full agreement with the thermodynamic linkage principle [45], but at constant MgPEP the A0:5 is much lower at low than at high fPEP. This explains the di¡erent e¡ects caused by the activator on the kinetics of saturation by MgPEP at high and low fPEP (Fig. 2) and, again, suggests competition between free phosphomycin and fPEP for binding to the allosteric site, in accordance with previous experimental evidence of a regulatory site for PEP [14], which is the allosteric Glc6P

BBAPRO 35674 14-7-98

140

C. Mu¨jica-Jime¨nez et al. / Biochimica et Biophysica Acta 1386 (1998) 132^144

Fig. 5. Kinetics of saturation of maize leaf PEPC by phosphomycin at pH 7.3. and at 0.11 (E) and 22.2 (a,b) mM fMg2‡ . MgPEP concentrations were 0.1 (b,E) or 2 mM (a), and fPEP concentrations were 5 (E), 0.025 (b), or 0.5 (a). Other assay conditions were as in Fig. 1. The points are the experimentally determined values, and the lines drawn through these points are calculated from the best ¢t of the data to Eq. 6 (E) or (9) (a,b). The initial velocities of the controls (v0 , no activator added) were 1.17 (E), 0.60 (b), and 5.95 (a) units/mg protein. The estimated values of A0:5 for free phosphomycin were: 1.49 þ 0.06 (E), 0.18 þ 0.02 (b), and 0.02 þ 0.00 (a) mM.

site [16,19,20]. Alternatively, this result may be indicative of a better binding to the allosteric site of the complex phosphomycin-Mg than of free phosphomycin, since at ¢xed MgPEP the concentration of fMg2‡ and fPEP are inversely related by a ¢xed factor and, therefore, higher levels of Mg-phosphomycin are present at low than at high fPEP. To elucidate which form of phosphomycin, the free species or the phosphomycin-Mg complex, is the true activator, we measured the enzyme activity at ¢xed MgPEP when free phosphomycin was varied and Mg-phosphomycin held constant, and vice versa. In order to observe clear increases in activity, we used a very low substrate concentration (0.1 mM MgPEP). The results, shown in Fig. 4, clearly demonstrate that free phosphomycin is the true activator of maize leaf PEPC, since a 5-fold increase in Mg-phosphomycin did not a¡ect the enzyme activity already increased by the presence of 0.1 mM free phosphomycin, while a 5-fold increase in free phosphomycin caused a further 2-fold increase. Similar results were obtained with Glc6P (data not shown), which, strongly support that the Glc6P allosteric-site binds the uncom-

plexed, free forms of the activators, contrary to what it was ¢rst proposed [32]. The kinetics of saturation of the enzyme by free phosphomycin, in a concentration range from zero to 10 mM, at 0.1 mM MgPEP exhibited deactivation at free phosphomycin concentrations above 1 mM when the concentration of fMg2‡ was high, 22.2 mM (Fig. 5, open and closed circles), but no deactivation was observed when the fMg2‡ concentration was low, 0.11 mM (Fig. 5, open squares). This result is consistent with inhibition by phosphomycin-Mg, since there is 20-times more Mg-phosphomycin in the former than in the latter conditions. At the low fMg2‡ concentration, the levels of this complex seem not to be high enough to compete with MgPEP for binding to the active site, and to produce kinetically signi¢cant levels of the complex active site-Mg-phosphomycin, even at the highest free phosphomycin concentration used, 10 mM. However, at the high fMg2‡ concentration, Mg-phosphomycin reaches concentrations as high as 50 mM, at 10 mM free phosphomycin, which clearly are able to compete

Fig. 6. Kinetics of saturation by MgPEP of maize leaf PEPC at pH 7.3 in the absence (a) or presence of 20% (v/v) glycerol (b) or 20% (v/v) glycerol plus 5 mM phosphomycin (F). Assays were run at pH 7.3, varying the concentrations of MgPEP and fPEP and keeping constant the concentration of fMg2‡ at 1 mM. Other assay conditions were as in Fig. 1. The points are the experimentally determined values, and the lines drawn through these points are calculated from the best ¢t of the data to Eq. 2 (a) or (3) (b,F). The estimated values of Km (MgPEP) were: 1.62 þ 0.18 (a), 0.31 þ 0.08 (b), and 0.08 þ 0.01 (F) mM.

BBAPRO 35674 14-7-98

C. Mu¨jica-Jime¨nez et al. / Biochimica et Biophysica Acta 1386 (1998) 132^144

141

Table 2 E¡ect of 20% (v/v) glycerol on the kinetics of saturation of maize leaf PEPC by phosphomycina Assay conditionsb pH 7.3 Control +Glycerol pH 8.3 Control +Glycerol

Kinetic parametersc Maximum activation (times)

A0:5 (WM)

n

17.2 þ 1.1 0.2 þ 0.0

497 þ 124 39 þ 8

1.26 þ 0.11 1

13.1 þ 0.2 0.4 þ 0.0

286 þ 13 27 þ 6

1.87 þ 0.13 1

a

Total phosphomycin concentrations were varied in a range from 0.025 to 5.0 mM. Assays were performed at 0.1 mM MgPEP, 1 mM fMg2‡ , and 0.55 mM fPEP. c Values þ S.E. were estimated by the best ¢t to Eq. 5 or Eq. 6. b

with the substrate causing a decrease in the maximum activation achieved by the activator. Accordingly, the deactivation observed was somewhat more pronounced at 0.1 than at 2 mM MgPEP (Fig. 5), with appKi (Mg-phosphomycin) values of 26.7 þ 3.7 and 44.8 þ 5.9 mM, respectively. These results are in full agreement with the those obtained when the kinetics of saturation by MgPEP were run at high fMg2‡ concentrations (Fig. 2B). Taken together, our ¢nding of the di¡erent forms of the activator bound to the allosteric and active sites of PEPC provides an explanation to previous reports [10,11,13] of phosphate esters behaving as activators and as inhibitors, which would depend on the relative concentrations of their free and metal-complexed forms under the conditions of the assay, and on the substrate concentration. However, given the low levels of fMg2‡ prevailing in vivo under any conceivable physiological condition [31] and the low stability constant of the Mg2‡ complexes of the known physiological activators of PEPC [32], binding of these activators to the active site will not be relevant under intracellular conditions.

the substrate observed in the absence of the cosolute at pH 7.3 but did not change the heterotropic e¡ects of phosphomycin on the kinetics of saturation by MgPEP (Fig. 6). In fact, the cosolute greatly increased the a¤nity of the enzyme for the activator, as suggested by the 10-fold reduction in the A0:5 values obtained at low MgPEP and fMg2‡ concentrations in the presence of the cosolute (Table 2). Glycerol also abolished the cooperative binding of phosphomycin both at pH 7.3 and 8.3. Thus, the e¡ects of glycerol on the kinetics of saturation by phosphomycin are similar to those on the kinetics of saturation by the substrate. At present we cannot conclude whether they are a consequence of a putative glycerol-induced conformational change that would a¡ect both the active and Glc6P allosteric sites, or only one of them, since the higher degree of saturation of the enzyme by either the substrate or the activator in the presence of the cosolute will lead to higher binding of the other through heterotropic e¡ects. Binding studies will be required to clarify this point.

3.4. E¡ect of glycerol on the activation of phosphoenolpyruvate carboxylase by phosphomycin

4. Discussion

It is known that glycerol reduces the appKm (PEP) of maize-leaf PEPC [6] and that its activating e¡ect is additive to that of Glc6P [46,47]. In agreement with these reports, we found that the inclusion of 20% (v/ v) glycerol in the assay medium caused an increase in the a¤nity of the enzyme for the substrate MgPEP, abolished the positive homotropic cooperativity of

The elucidation of the structural di¡erences that allow discrimination between PEP-analog inhibitors and activators of PEPC is of special relevance to understanding the regulation of the enzyme, since PEP itself act as a substrate and activator [16,19,20]. We report here that another PEP analog, phosphomycin, is a good activator of this enzyme. The chemical structure of phosphomycin is quite di¡erent from that of the other known activators which are also

BBAPRO 35674 14-7-98

142

C. Mu¨jica-Jime¨nez et al. / Biochimica et Biophysica Acta 1386 (1998) 132^144

PEP analogs [14^16], what make this compound potentially valuable in future studies of structure^function relationships of the Glc6P allosteric site. Comparing the chemical structures of the PEP analogs that behave as PEPC activators ([14^16], and this work) with those of the known PEP analogs that behave as PEPC inhibitors (see Table 1 of ref. [17]), it is clear that the activators are dianions while the inhibitors are trianions. On the other hand, our results indicate that while the Glc6P allosteric site binds the non-complexed form of the activator, the active site binds preferentially the metal-activator complex, in accordance with the preferential binding of the metal-substrate complex over the free substrate by the latter [16,20,26,27]. Therefore, it seems that there are some critical structural di¡erences between both sites, in spite of their apparent similarity. Thus, we propose that whether a given compound with the right chemical structure to bind to the active and allosteric sites would inhibit or activate the enzyme is, at least in part, the result of: (1) its ability to form complexes with magnesium; and (2) the structure of the resulting magnesium complex. The physiological relevance of the di¡erences in the form of substrate and activator bound to the active and Glc6P-allosteric sites is not totally understood at present, although it is clear that, under physiological conditions, these di¡erences would minimize the competition between substrate and activators for binding to the active site. One interesting result of this work is the ¢nding that phosphomycin caused increases in the apparent substrate inhibition of PEPC by MgPEP both at ¢xed fPEP (Fig. 2B) and fMg2‡ (not shown). Increased substrate inhibition was also observed in the presence of Glc6P (not shown), or glycerol (Fig. 6). At present, the mechanism underlying substrate inhibition of this enzyme is unknown, but inhibition might result from abortive binding of MgPEP to one of the enzyme^product complexes. Alternatively, if the free species, Mg2‡ or PEP, are uncompetitive or mixed inhibitors with respect to the complex MgPEP, an apparent substrate inhibition pattern will be observed under the conditions of our assay, since they are varied in a constant ratio to the substrate [44]. In the latter case, inhibition will be more pronounced as the [free species]/[MgPEP] ratio increases, as in fact was observed. Thus, at

5 mM fPEP, when the [fMg2‡ ]/[MgPEP] ratio is only 1.11, there was no substrate inhibition, at least in the MgPEP and fMg2‡ concentration range used in this study, while at 0.5 mM fPEP when this ratio is 11.1 there was clear inhibition (Fig. 2). Similar substrate inhibitions were observed at ¢xed fMg2‡ , when the [fPEP]/[MgPEP] ratios were higher than 5.55 (data not shown). These ¢ndings suggest that the species causing the apparent substrate inhibition are fPEP and/or fMg2‡ , rather than MgPEP. If this is true, the ligands of the Glc6P allosteric site or glycerol should increase the binding to the active site not only of the substrate MgPEP, but also of the free species PEP and Mg2‡ , to account for the increased substrate inhibition observed in their presence. Phosphomycin binds cooperatively at low substrate concentrations at both pH 7.3 and 8.3, with n values ranging from 1.26 to 1.87. The heterotropic e¡ects of phosphomycin on the kinetics of saturation for MgPEP were observed also at the two pH values, even in the presence of glycerol. These ¢ndings are interesting since there were no homotropic e¡ects for the substrate either at pH 8.3, or at pH 7.3 in the presence of glycerol. The uncoupling of the homotropic e¡ects for the substrate and the homotropic and heterotropic e¡ects for the activator argues against any model of two states that attempts to explain the allosteric behavior of maize leaf PEPC. However, to date, the two mechanisms proposed to underlie the allosteric properties observed in PEPC are oligomerization, i.e. conversion of the inactive dimer to the active tetramer, induced by the substrate [46] and concerted isomerization [48] according to the model developed by Monod et al. [49]. Regarding the ¢rst, it is known that formation or stabilization of the active tetrameric form of the enzyme accompanies activation by Glc6P [50] or glycerol [46]. These observations concerning activation and oligomerization have led to the supposition that changes in quaternary structure may be directly involved in the change in kinetic characteristics. Oligomerization may be su¤cient for activation, in which case phosphomycin would function by enhancing oligomerization, either by binding preferentially to low Km oligomers, or by binding to dimers and inducing a conformational change that produces oligomerization. However, oligomerization alone cannot be suf-

BBAPRO 35674 14-7-98

C. Mu¨jica-Jime¨nez et al. / Biochimica et Biophysica Acta 1386 (1998) 132^144

¢cient to cause the phosphomycin-induced allosteric activation, since we found signi¢cant phosphomycininduced decreases in Km (MgPEP) at pH 8.3 or in the presence of glycerol, when the enzyme does not exhibit any cooperativity in the binding of substrate and it is known to be predominantly tetrameric [46]. So, in addition to possibly a¡ecting the aggregation state of PEPC, phosphomycin appears to induce a putative conformational change that activates the carboxylase. This is in agreement with the recent ¢ndings of homotropic and heterotropic e¡ects in the tetrameric form of maize leaf PEPC [51]. Regarding the proposed concerted mechanism [48], our results indicate that the e¡ects of phosphomycin are di¡erent from those of a classical allosteric e¡ector that causes a transition between R and T states. The Monod model for allosterism [49] assumes that the allosteric e¡ectors merely shift the equilibrium between the R and T states, without changing the properties of these states. The uncoupling of the allosteric interactions of the substrate and activator reported here, together with the di¡erences in the degree of activation elicited by ligands of the same allosteric site, as are phosphomycin and Glc6P (Fig. 1B), and the summation of the e¡ects of two allosteric activators which bind to di¡erent allosteric sites, such as glycine and phosphomycin (Fig. 1C), show that this cannot be true for maize-leaf PEPC. In conclusion, whatever the mechanism is, to account for the results reported here and elsewhere [34,51] multiple conformational states of the C4 -PEPC should be postulated. Acknowledgements This work was supported by a grant to RAMC from the Direccio¨n General de Apoyo al Personal Acade¨mico de UNAM (DGAPA-IN 211694). A.C.M. was a recipient of a Consejo Nacional de Ciencia y Tecnolog|¨a (CONACYT) scholarship. We thank A. Tovar-Me¨ndez for carrying out preliminary work with phosphomycin. References [1] J.A. Jiao, R. Chollet, Arch. Biochem. Biophys. 269 (1990) 526^535.

143

[2] J. Vidal, R. Chollet, Trends Plant Sci. 2 (1997) 230^237. [3] J. Coombs, C.W. Baldry, C. Bucke, Planta 110 (1973) 95^ 107. [4] M. Stiborova, S. Leblova, Photosynthetica 19 (1985) 177^ 182. [5] T. Nishikido, H. Takanashi, Biochem. Biophys. Res. Commun. 53 (1973) 126^133. [6] K. Uedan, T. Sugiyama, Plant Physiol. 57 (1976) 906^910. [7] S.C. Huber, G.E. Edwards, Can. J. Bot. 53 (1975) 1925^ 1933. [8] M.H. O'Leary, Physiol. Veg. 21 (1983) 883^888. [9] V. Bandarian, W.J. Poehner, S.D. Grover, Plant Physiol. 100 (1992) 1411^1416. [10] H.D. Doncaster, R.C. Leegood, Plant Physiol. 84 (1987) 82^ 87. [11] E. Selinioti, G. Karabourniotis, Y. Manetas, N.A. Gavalas, J. Plant Physiol. 121 (1985) 353^360. [12] D. Lavergne, M.-L. Champigny, Physiol. Veg. 21 (1983) 1021^1029. [13] D.H. Gonza¨lez, A.A. Iglesias, C.S. Andreo, Biochem. J. 241 (1987) 543^548. [14] C.L.D. Jenkins, R.L.N. Harris, H.G. McFadden, Biochem. Int. 12 (1986) 619^625. [15] J.M. Peliska, M.H. O'Leary, Biochemistry 28 (1989) 1604^ 1611. [16] R. Rodr|¨guez-Sotres, R.A. Mun¬oz-Clares, Arch. Biochem. Biophys. 276 (1990) 80^190. [17] D.H. Gonza¨lez, C.S. Andreo, Trends Biol. Sci. 14 (1989) 24^ 27. [18] C.L.D. Jenkins, R.L.N. Harris, H.G. McFadden, Biochem. Int. 14 (1987) 219^226. [19] P. Rustin, C.R. Meyer, R.T. Wedding, J. Biol. Chem. 263 (1988) 17611^17614. [20] A. Tovar-Me¨ndez, R. Rodr|¨guez-Sotres, D.M. Lo¨pez-Valent|¨n, R.A. Mun¬oz-Clares, Biochem. J. 332 (1998) 633^642. [21] B.G. Christensen, W.J. Leanza, T.R. Beattie, A.A. Patchett, B.H. Arison, R.E. Ormond, F.A. Kuehl Jr., G. AlbersSchonberg, O. Jardetzky, Science 166 (1969) 123^125. [22] D. Hendlin, E.O. Stapley, M. Jackson, H. Wallick, A.K. Miller, L. Chaiet, F.M. Kahan, E.L. Foltz, H.B. Woodru¡, J.M. Mata, S. Herna¨ndez, S. Mochales, Science 166 (1969) 121^122. [23] F.M. Kahan, J.S. Kahan, P.J. Cassidy, H. Kropp, Ann. N.Y. Acad. Sci. 235 (1974) 364^385. [24] C. Wanke, N. Amrhein, Eur. J. Biochem. 218 (1993) 861^ 870. [25] J.W. Janc, M.H. O'Leary, W.W. Cleland, Biochemistry 31 (1992) 6421^6426. [26] S.K. Mukerji, Arch. Biochem. Biophys. 182 (1977) 352^359. [27] R.T. Wedding, P. Rustin, C.R. Meyer, M.K. Black, Plant Physiol. 88 (1988) 976^979. [28] F. Wold, C.E. Ballou, J. Biochem. Chem. 227 (1957) 301^ 312. [29] K.L. Manchester, Biochem. Biophys. Acta 630 (1980) 225^ 231. [30] T.E. Smith, Arch. Biochem. Biophys. 137 (1970) 512^522.

BBAPRO 35674 14-7-98

144

C. Mu¨jica-Jime¨nez et al. / Biochimica et Biophysica Acta 1386 (1998) 132^144

[31] Y. Yazaki, N. Asukagawa, Y. Ishikawa, E. Ohta, M. Sakata, Plant Cell Physiol. 29 (1988) 919^924. [32] W.J.O. O'Sullivan, G.W. Smithers, in: D.L. Purich (Ed.), Methods in Enzymology, Vol. 63, Academic Press, New York, 1979, pp. 294^336. [33] R.T. Wedding, M.K. Black, Plant Physiol. 87 (1988) 443^ 446. [34] A. Tovar-Me¨ndez, C. Mu¨jica-Jime¨nez, R.A. Mun¬oz-Clares, Biochim. Biophys. Acta 1337 (1996) 207^216. [35] Y.-H. Wang, S.M.G. Du¡, L. Lepiniec, C. Cre¨tin, G. Sarath, S.A. Condon, J. Vidal, P. Gadal, R. Chollet, J. Biol. Chem. 267 (1992) 16759^16762. [36] S.M.G. Du¡, C.S. Andreo, V. Pacquit, L. Lepiniec, G. Sarath, S.A. Condon, J. Vidal, P. Gadal, R. Chollet, Eur. J. Biochem. 228 (1995) 92^95. [37] M.M. Bradford, Anal. Biochem. 72 (1976) 248^254. [38] D.W. Marquardt, J. Soc. Ind. Appl. Math. 11 (1963) 431^ 441. [39] I.H. Segel, in: Enzyme Kinetics, Wiley, New York, 1975, pp. 227^272. [40] A.A. Iglesias, C.S. Andreo, Photosynth. Res. 5 (1984) 215^ 216.

[41] D.H. Gonza¨lez, A.A. Iglesias, C.S. Andreo, Arch. Biochem. Biophys. 245 (1986) 179^186. [42] J. Gillinta, S.D. Grover, Photosynth. Res. 45 (1995) 121^126. [43] R.T. Wedding, M.K. Black, C.R. Meyer, Plant Physiol. 90 (1989) 648^652. [44] I.H. Segel, in: Enzyme Kinetics, Wiley, New York, 1975, pp 147^151. [45] J. Wyman, in: Advances in C.B. An¢nsen, J.T. Edsall, M.L. Anson, F.M. Richards (Eds.), Protein Chemistry, Vol. 19, Academic Press, New York, 1964, pp. 223^286. [46] F.E. Podesta¨, C.S. Andreo, Plant Physiol. 90 (1989) 427^433. [47] R.T. Wedding, C.E. O'Brien, K. Kline, Plant Physiol. 104 (1994) 613^616. [48] N. Jawali, Arch. Biochem. Biophys. 277 (1990) 69^73. [49] J. Monod, J. Wyman, J.-P. Changeux, J. Mol. Biol. 12 (1965) 88^118. [50] C.R. Meyer, K.O. Willeford, R.T. Wedding, Arch. Biochem. Biophys. 288 (1991) 343^349. [51] A. Tovar-Me¨ndez, H. Yampara-Iquise, C. Mu¨jica-Jime¨nez, R.A. Mun¬oz-Clares, in: P. Mathis (Ed.), Photosynthesis: from Light to Biosphere, Vol. V, Kluwer, Dordrecht, 1995, pp. 155^158.

BBAPRO 35674 14-7-98