Evaluation by site-directed mutagenesis of active site amino acid residues of Anaerobiospirillum succiniciproducens phosphoenolpyruvate carboxykinase

Journal of Protein Chemistry, Vol. 21, No. 6, August 2002 (© 2002)

Evaluation by Site-Directed Mutagenesis of Active Site Amino Acid Residues of Anaerobiospirillum succiniciproducens Phosphoenolpyruvate Carboxykinase Ana María Jabalquinto,1 Maris Laivenieks,2 Fernando D. González-Nilo,1 Alejandro Yévenes,1 María Victoria Encinas,1 J. Gregory Zeikus,2 and Emilio Cardemil1,3 Received April 29, 2002

Anaerobiospirillum succiniciproducens phosphoenolpyruvate (PEP) carboxykinase catalyzes the reversible formation of oxaloacetate and adenosine triphosphate from PEP, adenosine diphosphate, and carbon dioxide, and uses Mn2⫹ as the activating metal ion. The enzyme is a monomer and presents 68% identity with Escherichia coli PEP carboxykinase. Comparison with the crystalline structure of homologous E. coli PEP carboxykinase [Tari, L. W., Matte, A., Goldie, H., and Delbaere, L. T. J. (1997). Nature Struct. Biol. 4, 990–994] suggests that His225, Asp262, Asp263, and Thr249 are located in the active site of the protein, interacting with manganese ions. In this work, these residues were individually changed to Gln (His225) or Asn. The mutated enzymes present 3–6 orders of magnitude lower values of Vmax/Km, indicating high catalytic relevance for these residues. The His225Gln mutant showed increased Km values for Mn2⫹ and PEP as compared with wild-type enzyme, suggesting a role of His225 in Mn2⫹ and PEP binding. From 1.5–1.6 Kcal/mol lower affinity for the 3⬘(2⬘)-O-(N-methylantraniloyl) derivative of adenosine diphosphate was observed for the His225Gln and Asp263Asn mutant A. succiniciproducens PEP carboxykinases, implying a role of His225 and Asp263 in nucleotide binding. KEY WORDS: Anaerobiospirillum succiniciproducens; active site residues; phosphoenolpyruvate carboxykinase; site-directed mutagenesis.

Phosphoenolpyruvate (PEP) carboxykinases (guanosine triphosphate/adenosine triphosphate (GTP/ATP); oxaloacetate carboxy-lyase (transphosphorylating), EC 4.1.1. 32/49) catalyze the nucleoside triphosphate-dependent reversible decarboxylation of oxaloacetate to yield PEP, CO2, and the corresponding nucleoside diphosphate, and they are found in all groups of organisms (Utter and Kolenbrander, 1972). Phosphoenolpyruvate carboxykinases have an absolute requirement for divalent cations for activity. Mixed-metal studies showed a dual role for cations in

these enzymes (Lee et al., 1981; Colombo et al., 1981). One cation, preferentially a transition metal, interacts with the enzyme in metal binding site 1 to elicit activation, while the second cation, in metal binding site 2, interacts with the nucleotide to serve as the metal– nucleotide substrate. The X-ray crystal structure of the ATP-dependent Escherichia coli and Trypanosoma cruzi PEP carboxykinases have been solved (Matte et al., 1996; Trapani et al., 2001), and recently Dunten et al. (2002) reported the crystal structure of the human cytosolic GTP-dependent carboxykinase. The E. coli PEP

1

4

Departamento de Ciencias Químicas, Facultad de Química y Biología, Universidad de Santiago de Chile, Casilla 40, Santiago 33, Chile. 2 Department of Biochemistry, Michigan State University, East Lansing, Michigan 48824. 3 To whom correspondence should be addressed. E-mail: ecardemi@ lauca.usach.cl

Abbreviations: MantADP, 3⬘ (2⬘)-O-(N-methylantraniloyl) derivative of adenosine diphosphate; MantATP, 3⬘ (2⬘)-O-(N-methylantraniloyl) derivative of adenosine triphosphate; OAA, oxaloacetic acid; PEP, phosphoenolpyruvate; PIPES, piperazine-N-N⬘-bis(2-ethanesulfonic acid); SDS–PAGE, sodium dodecyl sulfate–polyacrylamide gel electrophoresis.

393 0277-8033/02/0800-0393/0 © 2002 Plenum Publishing Corporation

394 carboxykinase structure shows that the polypeptide consists of a 275-residue N-terminal domain and a more compact 265-residue C-terminal domain, with the active site located within a cleft between the two domains (Matte et al., 1996). Upon binding ATP-Mg2⫹-oxalate, the E. coli enzyme undergoes a domain closure through a 20-degree rotation of the N- and C-terminal domains toward each other (Tari et al., 1996). Two divalent metal cation binding sites are associated with the active site, with a 5.2-Å separation between the two ions. In both sites the metal ions are coordinated in an octahedral geometry. Mn2⫹, at site 1, is bonded by one imidazole (the N␧-2 atom of His232), one carboxylate (Asp269), two water molecules, an oxygen from P␥ of ATP, and N␧ from Lys213. Mg2⫹, at site 2, is coordinated with three water molecules, two oxygen atoms from the ␤- and ␥phosphoryl groups of ATP, and O␥ of Thr255. Two of the water molecules coordinating Mg2⫹ are also hydrogen bonded to the carboxylate groups of Asp268 and Asp269 (Tari et al., 1997). The structure of the active site region of the T. cruzi and human PEP carboxykinases shows remarkable similarities to the E. coli enzyme. Despite the detailed structural knowledge of these enzymes, very few reports deal with the evaluation of active site residues of PEP carboxykinases (Krautwurst et al., 1998; Llanos et al., 2001). Anaerobiospirillum succiniciproducens PEP carboxykinase is 68% identical to the E. coli enzyme, suggesting important structural similarities between the two carboxykinases. We are developing a project aimed to characterize the A. succiniciproducens enzyme, a carboxykinase that functions physiologically for the catabolic synthesis of succinate (Laivenieks et al., 1997; Jabalquinto et al., 1999) and not for gluconeogenesis, as other well-studied PEP carboxykinases (Matte et al., 1997). In this work we evaluate the catalytic relevance of His225, Asp262, Asp263, and Thr249 in the A. succiniciproducens enzyme. These residues are equivalent to His232, Asp268, Asp269, and Thr255 in E. coli PEP carboxykinase, where they appear as ligands of the metal ions at metal binding sites 1 and 2 of the enzyme. Our results support the proposal that these residues are part of the equivalent metal binding sites in A. succiniciproducens PEP carboxykinase, and allow a better understanding of the roles these residues play in catalysis in this enzyme.

1. EXPERIMENTAL PROCEDURES Lactate dehydrogenase, reduced nicotinamide adenine dinucleotide, MnCl2, adenosine diphosphate (ADP), and PEP were from Sigma (St. Louis, MO, U.S.A.). Ox-

Jabalquinto et al. aloacetic acid (OAA) and restriction endonucleases were from Boehringer Mannheim (Indianapolis, IN, U.S.A.), Pfu and Pfu Turbo DNA polymerases were from Stratagene (La Jolla, CA, U.S.A.). Ni-NTA agarose was purchased from Qiagen (Valencia, CA, U.S.A.). Thermo Sequenase Radiolabeled Terminator Cycle sequencing kit was from USB Corporation (Cleveland, OH, U.S.A.). Benchmark™ Protein Ladder was from GIBCO BRL Life Technologies (Rockville, MA, U.S.A.). Oligonucleotides were synthesized by the Michigan State University Biochemistry and Molecular Biology Department Macromolecular Structure Facility. All other reagents were of the highest purity commercially available. 3⬘ (2⬘)O-(N-Methylantraniloyl) derivative of ADP (MantADP derivative)4 was synthesized as described by Carrasco et al. (1998). Mutagenesis and expression of recombinant wildtype and mutant enzymes. All DNA manipulations were performed using established methods and protocols (Sambrook et al., 1989; Ausubel et al., 1993). Point mutations were introduced into the A. succiniciproducens pckA gene by polymerase chain reaction using the Stratagene QuickChange™ Site-Directed Mutagenesis procedure with plasmid pPCK1 (Laivenieks et al., 1997) as the template. Mutated genes were sequenced and subcloned into the SacI–SalI sites of plasmid pProPC1, producing plasmids pProH225Q, pProD262N, pProD263N, and pProT249N. E. coli JM109 (Stratagene) was used for plasmid DNA purification and maintenance, and E. coli PB25 (Jabalquinto et al., 1999) was used for protein expression and purification. Plasmid-containing strains were grown in Luria-Bertani media containing 100 ␮g/ml ampicillin, chloramphenicol (10 ␮g/ml), and kanamycin (10 ␮g/ml) (E. coli PB25). Recombinant enzyme expression was induced by 0.6 mM isopropyl-␤-D-thiogalactopyranoside at 37°C for 2.5 hr. Recombinant enzymes were purified as described (Laivenieks et al., 1997). Assays and kinetic studies. PEP carboxykinase was assayed in the PEP carboxylation direction at 30°C in a 1-mL reaction mixture containing 100 mM piperazine-NN⬘-bis(2-ethanesulfonic acid) (PIPES) buffer (pH 6.5), as described (Jabalquinto et al., 1999). One unit of enzyme activity is defined as the amount of enzyme that produces 1 ␮mol of OAA min⫺1. To obtain the kinetic parameters, enzyme activity was measured as the concentration of either substrate was varied while keeping the concentration of the other substrates at saturating levels. ADP was varied between 0.7 and 8 Km, PEP between 0.5 and 5 Km, and HCO⫺3 between 0.3 and 3 Km. Maximal velocity and Km values were determined by fitting the data to the Michaelis-Menten equation with the Microcal™ Origin™ program. Km values reported for HCO⫺3 represent the

Site-Directed Mutagenesis

395

total concentration of potassium bicarbonate added. For the Km values for Mn2⫹, free cation concentration was calculated with the GEOCHEM-PC V.2.0 program (Parker et al., 1995), using Kd values of 8.10⫺5 M for the MnADP2⫺ complex and 1.8 ⫻ 10⫺3 M for the MnPEP complex (Martel and Smith, 1998). The inhibitory effect of oxalate against PEP was determined for the recombinant wild-type and His225Gln PEP carboxykinases. For the wild-type enzyme, oxalate fixed concentrations were 0, 25, 50, or 100 ␮M, while the PEP concentration varied from 0.085–2.44 mM. For the His225Gln mutant, oxalate fixed concentrations were 0, 2.5, 5.0, and 10 mM, while the PEP concentration varied from 1.21–9.72 mM. All other conditions were the same as for the standard assay. Initial velocity data were fit with the EZ-FIT program (Perrella, 1988) to models for competitive, noncompetitive, and mixed noncompetitive inhibitions; the best fit was obtained for competitive inhibition. Interaction of A. succiniciproducens PEP carboxykinases with MantADP. MantADP binding was measured by the quenching of intrinsic protein fluorescence. Spectra were recorded at 22°C in a Spex 1681 Fluorolog spectrofluorimeter, with 1.25 nm slits for excitation and emission. The nucleotides were added stepwise in small volumes to 2–4 ␮M enzyme solutions in 50 mM PIPES buffer (pH 7.0) and 1 mM MnCl2. The maximum concentrations of ADP or MantADP added did not exceed 100 ␮M, thus ensuring that they were at least 95% present as the corresponding metal-nucleotide complexes. Upon excitation at 295 nm, the intrinsic fluorescence spectra were scanned from 310–480 nm. Corrections were made for the inner filter effect of the nucleotide analogue under the same conditions according to Lehrer and Leavis (1978). After correction for dilution by titrant, curve fitting of concentration-dependent changes in fluorescence was done with the Microcal™ Origin™ program using Eq. (1): F ⫽ Fmax ⫺ (Fmax ⫺ Fmin)[(Et ⫹ L ⫹ Kdiss) ⫺ ((Et ⫹ L ⫹ Kdiss)2 ⫺ 4EtL)1/2]/2Et

[1]

where F is the relative fluorescence intensity, Fmax the relative fluorescence intensity at the beginning of the titration, Fmin the fluorescence intensity at saturating concentration of ligand L, Et the enzyme concentration, and Kdiss the dissociation constant of the enzyme-ligand complex. For competitive displacement of ADP by MantADP from the enzyme-ADP complex, Eq. (2) was used (Segel, 1995; Faller, 1989): Kdiss(app) ⫽ Kdiss(1 ⫹ ADP/KdissADP)

[2]

where Kdiss(app) is the apparent dissociation constant of the enzyme-MantADP complex in the presence of a certain ADP concentration, Kdiss the dissociation constant of the enzyme-MantADP complex in the absence of ADP, and KdissADP the dissociation constant for the enzymeADP complex. Circular dichroism spectroscopy. Circular dichroism (CD) spectra were recorded in an Aviv 26A DS CD spectrometer in 20 mM potassium phosphate buffer (pH 6.8) at 25°C using 0.1 cm path cells and a protein concentration of ⬃0.15 mg/mL. Computer-assisted three-dimensional homology modeling. The programs InsightII, Homology, and Discover 972 (Biosym/MSI) were used on O2 SGI workstations to build homology-based three-dimensional models of A. succiniciproducens PEP carboxykinase. The reference protein used as a template to follow the local degree of structure conservation or divergence was the E. coli PEP carboxykinase-ATP-pyruvate-Mg2⫹-Mn2⫹ complex (1AQ2) (Tari et al., 1997). All calculations were done with Discover_3 (MSI) and force field ESFF (MSI), which has all parameters needed for the octahedral Mn2⫹ coordination and amino acids. This program was also employed for energy minimization and molecular dynamics. The sequence alignment of E. coli and A. succiniciproducens PEP carboxykinases was made using the module BestFit of GCG (Genetic Computer Group) Wisconsin Package, SeqWeb2 (Henikoff and Henikoff, 1992). The matrix used for the sequence comparison was BLOSUM62 with a set gap creation penalty of 8 and a set gap extension penalty of 2. The alignment had an average match of 2.778 and an average mismatch of ⫺2.248, with only three small gaps. The two enzyme sequences showed 76% similarity and 68% identity. Mn2⫹ (metal binding site 1) was coordinated to two water molecules, an oxygen atom from P␥ of ATP, N2 from His225, an oxygen atom from the side chain of Asp263, and N␧ from Lys206. In metal binding site 2, Mg2⫹ was replaced by Mn2⫹, which was coordinated to three water molecules, two oxygen atoms from P␤ and P␥ of ATP, and the oxygen of the hydroxyl group of Thr249. The initial structure was relaxed using a distance-dependent dielectric constant of 80. The relaxed structure was used to replace pyruvate with PEP. To build the PEP structure, bonds were reorganized between the phosphate groups of ATP and the pyruvate molecule. Discover_3 was then employed to relax the protein structure using energy minimization and simulated annealing based on molecular dynamics. Conserved regions were restrained in the first 100 ps; then they were slowly unrestrained for 100 ps. The final relaxation was 300 ps. Nonbonding interactions were calculated using the cell multipole method for van

396

Jabalquinto et al.

der Waals and Coulombic interactions. The homology model of the A. succiniciproducens enzyme was validated using the PROCHECK program (Laskowski et al., 1993; Morris et al., 1992). An overall average G factor of ⫺0.28 showed that the structure was well stabilized. The Ramachandran plot showed 87.9% of the amino acids within the allowed zone.

tered enzymes were compared with the wild-type spectrum. The spectra (not shown) were found to be essentially the same, with a negative peak at 208 nm and a negative shoulder at 222 nm. The ratio of the mean residue ellipticity at 208 and 222 nm was between 0.74 and 0.75, indicating that the secondary structure of A. succiniciproducens PEP carboxykinase is not significantly altered by mutation at any of the above-indicated positions. Alterations in the tertiary structure were analyzed through the intrinsic fluorescence spectrum of the enzymes, which have a total of nine Trp residues at positions 77, 78, 95, 136, 199, 267, 295, 437, and 500 (Laivenieks et al., 1997). No alteration in the position of the emission or in the fluorescence intensity was detected, indicating that these mutations do not change the microenvironment of the Trp residues. Kinetic parameters of the wild-type and mutant enzymes. The kinetic parameters of all mutated enzymes are summarized in Table 1. The altered PEP carboxykinases showed significant alterations in Vmax, ranging from 270-fold reduction in the His225Gln mutant to 29,000-fold reduction in Asp262Asn PEP carboxykinase. Table 1 also shows that the His225Gln mutant has increased Km values for Mn2⫹ and PEP (27- and 13fold, respectively). A slight increase in Km for Mn2⫹ is also seen for the Asp263Asn mutant enzyme. Fourfold and eightfold increases in Km for MnADP are seen in the Asp262Asn and Thr240Asn mutant enzymes. All other Km values are within the same order of magnitude as those of the wild-type enzyme. Using the data shown in Table 1, it is possible to calculate the effect of the mutations in the Vmax/Km parameter, and use it as a measure of the relative catalytic efficiency of wild-type and mutant PEP carboxykinases. The values obtained (not shown) indicate 10⫺3–10⫺6 lower values for mutant PEP carboxykinases as compared with those of the wild-type enzyme. The lower Vmax and Vmax/Km values obtained for

2. RESULTS Expression and purification of recombinant wildtype and mutant A. succiniciproducens PEP carboxykinases. To study the catalytic role of active site residues His225, Asp262, Asp263, and Thr249, these residues were mutated to Gln (His225) or Asn. Wild-type and mutant PEP carboxykinases were expressed and purified from the pyruvate kinase deficient E. coli strain PB25. Sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS–PAGE) gel scanning data indicated that the overexpressed PEP carboxykinases represented ⬃10%–20% of the total crude cell extract protein. Yields of purified enzymes were in the range 70–150 mg/L of induced culture. All PEP carboxykinases had an N-terminal end fused His-tag, which facilitated the isolation of highly purified proteins in a single step (Laivenieks et al., 1997). Using 5.0 ␮g of protein per lane on SDS–PAGE, PEP carboxykinases were found to be homogeneous, and only at 2–3-fold higher loading levels did trace amounts of two additional protein bands became visible (not shown). The presence of the 22 amino acids His-tag and spacer sequence causes no alteration in the kinetic parameters of wild-type A. succiniciproducens PEP carboxykinase (Laivenieks et al., 1997). To examine whether the mutations at positions 225, 262, 263, and 249 disrupted the secondary structure of the enzyme, the far-ultraviolet CD spectra of the four al-

Table 1. Apparent Kinetic Parameters for Wild-Type and Mutant A. succiniciproducens PEP Carboxykinasesa Km Enzyme Wild-type His225Gln Asp262Asn Asp263Asn Thr249Asn a

Mn2⫹ (␮M) 48 ⫾ 4 1284 ⫾ 298 Nd 186 ⫾ 44 100 ⫾ 7

MnADP (␮M) 77 ⫾ 7 70 ⫾ 2 316 ⫾ 71 54 ⫾ 6 650 ⫾ 80

PEP (␮M) 219 2830 418 425 93

⫾ 20 ⫾ 420 ⫾ 51 ⫾ 57 ⫾8

HCO⫺3 (mM) 33 16 25 35 23

⫾5 ⫾1 ⫾6 ⫾6 ⫾2

Vmax (␮mol min⫺1 mg1) 79 0.29 0.0027 0.0080 0.084

⫾2 ⫾ 0.08 ⫾ 0.0002 ⫾ 0.0006 ⫾ 0.002

Kinetic constants were determined as indicated in Section 1, Experimental Procedures. Values given are the mean ⫾ standard deviation. Nd, Not determined.

Site-Directed Mutagenesis the mutant enzymes show that in A. succiniciproducens PEP carboxykinase, His225, Asp262, Asp263, and Thr249 are important residues for the catalytic efficiency of the enzyme. On the other hand, the Km values reported indicate that Asp262 and Thr249 are of relevance for the kinetic affinity of the enzyme for the metal-nucleotide complex, and suggest that His225 might participate in Mn2⫹ and PEP binding. To test further for the possible alteration in the PEP binding characteristics in the His225Gln mutant, the inhibitory properties of oxalate were determined. Oxalate is a structural analogue of enolpyruvate, the postulated reaction intermediate of PEP carboxykinase, and it has been reported to be either a competitive (Ash et al., 1990) or noncompetitive (Guidinger and Nowak, 1990) inhibitor of chicken liver PEP carboxykinase with respect to PEP. Recently, oxalate was found to be a mixed-type inhibitor against PEP in Mycobacterium smegmatis PEP carboxykinase (Mukhopadhyay et al., 2001). We have found (results not shown) that oxalate is a competitive inhibitor against PEP, with a Ki of 12 ⫾ 2 ␮M for the wild-type A. succiniciproducens PEP carboxykinase and of 2.7 ⫾ 0.7 mM for the 225Gln enzyme. The lower kinetic binding affinity of oxalate to His225Gln PEP carboxykinase as compared with the wild-type enzyme supports the involvement of His225 in the PEP binding site. Interaction of wild-type and altered A. succiniciproducens PEP carboxykinases with MantADP. The Km values for ADPMn of the mutant A. succiniciproducens PEP carboxykinases indicate reduced kinetic binding affinity for the Asp262Asn and Thr249Asn enzymes (Table 1). To obtain a better measurement of the binding affinity for the nucleotide of wild-type and mutated A. succiniciproducens PEP carboxykinases, the dissociation constants of the corresponding enzyme-MantADP complexes were measured. Binding of ADP or ATP to A. succiniciproducens PEP carboxykinase is not accompanied by changes of the intrinsic fluorescence of the protein; hence, ADP and ATP could not be used in fluorescence studies of nucleotide interactions with the enzyme. Binding of MantADP to proteins can be evaluated through the increase in the fluorophore emission of ⬃440 nm, by quenching of the intrinsic tryptophan fluorescence, or by the increase in emission of the nucleotide analogue if energy transfer from Trp residues takes place (Gonzalo et al., 2000). MantADP is a substrate for A. succiniciproducens PEP carboxykinase, which displays a specific activity of 0.22 ␮mol min⫺1n mg⫺1 at 0.20 mM nucleotide analogue in the PEP carboxylation direction. A kinetic analysis for MantADP was not done because the methylantraniloyl moiety of this substrate analogue absorbs at

397 340 nm, thus interfering with the activity assay at higher concentrations. MantADP binding to A. succiniciproducens PEP carboxykinase led to ⬍30% increase in fluorescence intensity with respect to the fluorescence of MantADP in water, and the emission maximum remained unaltered at 442 nm. These results indicate that the microenvironment sensed by the protein-bound fluorophore is very similar to that of water. This result is as expected from the position of the nucleotide ribose hydroxyls in the complex of the E. coli PEP carboxykinase with ATP, where they point toward the solvent, away from the protein structure (Tari et al., 1997). Modeling of similar complexes between MantADP and PEP carboxykinases from S. cerevisiae (Carrasco et al., 1998) or A. succiniciproducens (results not shown) indicate a solvent-exposed position for the N-methylanthraniloyl fluorophore. On the other hand, considering that at saturating levels MantADP produced ⬃60% quenching of the intrinsic fluorescence emission of the protein, we decided to employ fluorescence quenching to measure the binding of the nucleotide analogue to A. succiniciproducens PEP carboxykinase. Fitting of the experimental results to Eq. (1) are shown in Fig. 1A, both for the enzyme alone and for the enzyme in the presence of varying concentrations of ADP. The presence of ADP in the titration media decreased the apparent affinity of the enzyme for the analogue, because Kdiss for the enzymeMantADP complex increased from 1.7 ⫾ 0.5 ␮M in the absence of ADP to 51 ⫾ 5 ␮M in the presence of 69.6 ␮M ADP. Assuming that binding of ADP and MantADP to A. succiniciproducens PEP carboxykinase is mutually exclusive, Eq. (2) could be employed to calculate the dissociation constant for the enzyme-ADPMn complex (KdissADPMn). Data of Kapp as a function of ADP concentration fitted to Eq. (2) (Fig. 1B) yield KdissADPMn = 7.1 ⫾ 1.4 ␮M. Binding of MantADP to mutant A. succiniciproducens PEP carboxykinases was also measured, and it can be seen (Table 2) that decreased binding affinity for MantADP is seen only for His225Gln and Asp263Asn mutated enzymes.

3. DISCUSSION In this work, site-directed mutagenesis has been used to evaluate the catalytic properties of His225, Asp262, Asp263, and Thr249 in A. succiniciproducens PEP carboxykinase. To this purpose these amino acids, corresponding to His232, Asp268, Asp269, and Thr255 in the homologous E. coli enzyme (Matte et al., 1996), were changed to Gln (His225) or Asn (Asp262, Asp263, and Thr249).

398

Jabalquinto et al. Table 2. Dissociation Constants for Wild-Type and Mutant A. succiniciproducens PEP Carboxykinase-MantADP Complexes a Enzyme

Kdiss (␮M)

Wild-type His225Gln Asp262Asn Asp263Asn Thr249Asn

1.7 ⫾ 0.5 23 ⫾ 5 5 ⫾ 0.5 22 ⫾ 2 0.9 ⫾ 0.2

a

Fig. 1. A, Quenching of A. succiniciproducens PEP carboxykinase intrinsic fluorescence as a function of MantADP concentration, at various ADP concentrations. The enzyme (3 ␮M) was incubated with 1 mM MnCl2 in the absence (䊉) or presence of 1.74 ␮M (ⵧ), 17.4 ␮M (䉱), 34.8 ␮M (䉭), and 69.6 ␮M (䡲) ADP. All other conditions as described in the text. Lines are best fits to Eq. (1). B, Relationship between the dissociation constant of the A. succiniciproducens PEP carboxykinase-MantADP complex in the presence of increasing ADP concentrations. Dissociation constants obtained from Fig. 1A (Kdiss(app)) are plotted against the ADP concentration. The straight line is the best fit to Eq. (2) and gives a dissociation constant for the enzyme-ADP complex (KdissADP) of 7.1 ⫾ 1.4 ␮M.

Dissociation constants were obtained from data similar to that shown in Fig. 1A and fitted to Eq. (1). Values given are the media ⫾ standard deviation for at least three independent measurements.

Altered enzymes showed significantly lower Vmax and Vmax/Km values (Table 1), indicating catalytic roles for these residues in A. succiniciproducens PEP carboxykinase. The His225Gln and Asp263Asn A. succiniciproducens PEP carboxykinases had increased Km values for Mn2⫹ compared with that of the wild-type. Furthermore, a notable increase in Km for PEP was also evident in the His225Gln mutant. These results suggest that His225 and Asp263 are involved in Mn2⫹ binding and that alteration of His225 also affect PEP binding. These results agree with recent binding experiments carried out with homologous S. cerevisiae PEP carboxykinase, which showed two orders of magnitude decrease in Kdiss for the enzyme-Mn2⫹ complex when the equivalent His233 was changed to Gln (Krautwurst et al., 2001). Based on these considerations, it is possible that the increased Km value for Mn2⫹ in the His225Gln PEP carboxykinase reflects a lower affinity of the mutated protein for the metal ion. The increased Km for PEP in the His225Gln PEP carboxykinase correlates with the increased Ki for oxalate determined in the same enzyme, and suggests that the mutant enzyme has altered PEP and oxalate binding properties. This alteration in the PEP binding site of the enzyme can be interpreted using data from Tari et al. (1997). These data indicate that in the ATP-pyruvate-Mg2⫹-Mn2⫹ complex of E. coli PEP carboxykinase, one of the oxygen atoms of P␥ of ATP is linked to Mn2⫹ through its first coordination sphere. If such a coordination is maintained upon the phosphoryl transfer step, the enzyme-bound Mn2⫹ would be a ligand for PEP. Accordingly, any alteration in Mn2⫹ binding would affect PEP binding. Supporting the proposed interaction of Mn2⫹ and PEP, data obtained with both GTPand ATP-dependent PEP carboxykinases indicate that Mn2⫹ facilitates PEP binding (Hebda and Nowak, 1982; Duffy and Nowak, 1985). Recently, Dunten et al. (2002) have obtained the crystal structure of GTP-dependent human cytosolic PEP carboxykinase in a complex with PEP and Mn2⫹. In that structure, two oxygen atoms

Site-Directed Mutagenesis linked to P of PEP appear to interact with Mn2⫹, each through the intervening presence of a water molecule. This last structure supports earlier data from 31P-nuclear magnetic resonance imaging obtained with GTP-dependent chicken liver PEP carboxykinase (Duffy and Nowak, 1985). These data indicate that the distance between Mn2⫹ at site 1 and 31P of PEP in the enzyme-PEP-Mn2⫹ complex is 7.38 ⫾ 0.48 Å, implying that in this case coordination between PEP and Mn2⫹ is not within the inner sphere coordination region of the metal ion. Figure 2 depicts a modeled structure of the A. succiniciproducens PEP carboxykinase-ADP-PEP-Mn2⫹ complex, which is based on the ATP-pyruvate-Mg2⫹-Mn2⫹ complex of E. coli PEP carboxykinase (Tari et al., 1997), and which assumes that an oxygen atom of the PEP phosphate group is a first sphere coordination ligand for Mn2⫹. In this model, Lys206, His225, and Asp263 appear as first sphere coordination ligands of Mn2⫹ at site 1. The Asp262Asn and Thr249Asn mutant PEP carboxykinases showed greatly diminished Vmax and increased KmADPMn values, indicating that mutations affect both catalysis and the kinetic affinity of the metalnucleotide complex. This is expected, considering that Asp262 and Thr249 are involved in metal binding (Fig. 2). An estimation of the MnADP binding characteristics of the mutant enzymes was obtained by using MantADP. The fluorescent analogues MantADP and 3⬘ (2⬘)-O-(Nmethylantraniloyl) derivative of ATP (MantATP derivative) behave as substrate analogues for a number of nucleotide-using enzymes (Ni et al., 2000; Vertommen

Fig. 2. Stereodiagram of the homology model of A. succiniciproducens PEP carboxykinase-ADP-PEP-Mn2⫹ complex. The model was obtained using as reference the E. coli PEP carboxykinase-ATPpyruvate-Mn2⫹-Mg2⫹ complex (Tari et al., 1997). To build the PEP structure, bonds were reorganized between the phosphate groups of Mn2⫹ ions at metal sites 1 and 2 are shown as spheres.

399 et al., 1996; Jault et al., 2000), including S. cerevisiae PEP carboxykinase (Carrasco et al., 1998). MantADP is also a substrate for A. succiniciproducens PEP carboxykinase, and therefore, MantADP is an appropriate substrate to evaluate the nucleotide binding characteristics of this enzyme. Additional evidence of the binding specificity for MantADP was obtained through the demonstration of competitive binding of ADP and MantADP (Fig. 1B). The calculated dissociation constant for the enzyme-MnADP complex of 7.1 ⫾ 1.4 ␮M is in the range of the dissociation constant for the MgADP complex of E. coli PEP carboxykinase (15 ␮M) (Encinas et al., 1993) and the MnIDP complex of chicken liver PEP carboxykinase (7.5 ⫾ 2.5 ␮M) (Hebda and Nowak, 1982). The lower dissociation constant for MantADP of wild-type A. succiniciproducens PEP carboxykinase was 1.7 ⫾ 0.5 ␮M, indicating a somewhat higher affinity of this enzyme for the fluorescent analogue than for the nucleotide substrate, a situation also reported for other enzymes (Morris et al., 1992; Ni et al., 2000). Our data (Table 2) show a loss of 1.5–1.6 Kcal/ mol in the binding affinity for MantADP of the His225 or Asp263 mutated enzymes with respect to the wild-type, while no significant change is detected for the Asp262 and Thr249 altered enzymes. These results suggest that His225 and Asp263 are required for appropriate binding of ADP, a conclusion that is not apparent from just looking at the KmADPMn values. The decreased MantADP binding affinity in the His225Gln and Asp263Asn mutants can be understood considering that in these mutants lower Mn2⫹ binding capacity is expected, which could alter the electrostatic potential of the active site and thus affect the interaction of the nucleotide phosphoryl groups. The lower binding affinity could also be due to slight conformational changes in the nucleotide binding site, although none of the mutants showed alterations in the CD or intrinsic fluorescence spectra. On the other hand, mutation of Asp262 or Thr249 to Asn has a significant effect on the kinetic affinity of MnADP; however, the MantADP binding characteristics are almost unaltered. The latter is somewhat unexpected, considering that the modeled structure in Fig. 2 shows that Asp262 and Thr249 are second- and first-sphere ligands to the metal in the MnADP complex, respectively. These results show that proper interactions of Asp262 and Thr459 with the metal in the metal-nucleotide complex are required for catalysis but not for substrate binding in A. succiniciproducens PEP carboxykinase. ACKNOWLEDGMENTS We thank Dr. Mauricio Escudey for access to the GEOCHEM-PC V.2.0 program. Computer resources

400 were provided in part by the National Cancer Institute Frederick Biomedical Supercomputer Center through the collaboration of Dr. Raul Cachau. CD spectroscopy data were obtained at the University of Wisconsin-Madison Biophysical Instrumentation Facility (UW-Madison and BIR-9512577-NSF grants). This work was supported by the U.S. Department of Energy under research grants DE-FG02-93ER20108, FONDECYT 1000756, and DICYT-USACH 9941JL/020141GN. REFERENCES Ash, D. E., Emig, F. A., Chowdhury, S. A., Satoh, Y., and Schramm, V. L. (1990). J. Biol. Chem. 265: 7377–7384. Ausubel, F. M., Brent, R., Kingston, R. E., Moore, D. D., Seidman, J. G., Smith, J. A., et al. eds. (1993). Current Protocols in Molecular Biology, Greene Publishing and Wiley-Interscience, New York. Carrasco, L., González, F. D., and Cardemil, E. (1998). Biochim. Biophys. Acta 1429: 93–101. Colombo, G., Carlson, G. M., and Lardy, H. (1981). Biochemistry 20: 2749–2757. Duffy, T. H., and Nowak, T. (1985). Biochemistry 24: 1152–1160. Dunten, P., Belunis, C., Crowther, R., Hollfelder, K., Kammlott, U., Levin, W., et al. (2002). J. Mol. Biol. 316: 257–264. Encinas, M. V., Rojas, M. C., Goldie, H., and Cardemil, E. (1993). Biochim. Biophys. Acta 1162: 195–202. Faller, L. D. (1989). Biochemistry 28: 6771–6778. Gonzalo, P., Sontag, B., Lavergne, J. P., Jault, J. M., and Reboud, J. P. (2000). Biochemistry 39: 13558–13564. Guidinger, P. F., and Nowak, T. (1990). Arch. Biochem. Biophys. 278: 131–141. Hebda, C. A., and Nowak, T. (1982). J. Biol. Chem. 257: 5515–5522. Henikoff, S., and Henikoff, J. G. (1992). Proc. Natl. Acad. Sci. USA 89: 10915–10919. Jabalquinto, A. M., Laivenieks, M., Zeikus, J. G., and Cardemil, E. (1999). J. Protein Chem. 18: 659–664. Jault, J. M., Fieulaine, S., Nessler, S., Gonzalo, P., Di Pietro, A., Deutscher, J., et al., A. (2000). J. Biol. Chem. 275: 1773–1780. Krautwurst, H., Bazaes, S., González, F. D., Jabalquinto, A. M., Frey, P. A., and Cardemil, E. (1998). Biochemistry 37: 6295–6302.

Jabalquinto et al. Krautwurst, H., Roschzttardtz, H., González-Nilo, F. D., Bazaes, S., Nowak, T., and Cardemil, E. (2001). Protein Sci. 10: R-325. Laivenieks, M., Vieille, C., and Zeikus, J. G. (1997). Appl. Environ. Microbiol. 63: 2273–2280. Laskowski, R. A., MacArthur, M. W., Moss, M. D., and Thornton, J. M. (1993). J. Appl. Crystallogr. 26: 283–291. Lee, M. H., Hebda, C. A., and Nowak, T. (1981). J. Biol. Chem. 256: 12793–12801. Lehrer, S. S., and Leavis, P. C. (1978). Methods Enzymol. 49: 222–236. Llanos, L., Briones, R., Yévenes, A., González-Nilo, F. D., Frey, P. A., and Cardemil, E. (2001). FEBS Lett. 493: 1–5. Martel, A. E., and Smith, R. M. (1998). NIST Critically Selected Stability Constants of Metal Complexes. NIST standard references database 46 version 5.0. Matte, A., Goldie, H., Sweet, R. M., and Delbaere, L. T. J. (1996). J. Mol. Biol. 256: 126–143. Matte, A., Tari, L. W., Goldie, H., and Delbaere, L. T. J. (1997). J. Biol. Chem. 272: 8105–8108. Morris, A. L., MacArthur, M. W., Hutchinson, E. G., and Thornton, J. M. (1992). Proteins 12: 345–364. Mukhopadhyay, B., Concar, E. M., and Wolfe, R. S. (2001). J. Biol. Chem. 276: 16137–16145. Ni, D. Q., Shaffer, J., and Adams, J. A. (2000). Protein Sci. 9: 1818– 1827. Parker, D. R., Norvell, W. A., and Chaney, R. L. (1995). GEOCHEMPC: A chemical speciation program for IBM and compatible personal computers. In: Loeppert, R. H., et al. (eds.), Chemical Equilibrium and Reaction Models, SSSA Spec. Publ. No. 42, American Society of Agronomy, Madison, WI. Perrella, F. W. (1988). Anal. Biochem. 174: 437– 447. Sambrook, J., Fritch, E. F., and Maniatis, T. (1989). In: Nolan, C. (ed.), Molecular Cloning: A Laboratory Manual (2nd Ed.), Cold Spring Harbor Laboratory Press, Cold Spring Harbor, New York. Segel, I. H. (1995). Enzyme Kinetics, John Wiley and Sons, New York. Tari, L. W., Matte, A., Pugazhenthi, U., Goldie, H., and Delbaere, L. T. (1996). Nat. Struct. Biol. 3: 355–363. Tari, L. W., Matte, A., Goldie, H., and Delbaere, L. T. (1997). Nat. Struct. Biol. 4: 990–994. Trapani, S., Linss, J., Goldenberg, S., Fischer, H., Craievich, A. F., and Oliva, G. (2001). J. Mol. Biol. 313: 1059–1072. Utter, M. F., and Kolenbrander, H. M. (1972). The Enzymes (3rd Ed., Vol. 6), Academic Press, New York, pp. 117–168. Vertommen, D., Bertrand, L., Sontag, B., Di Pietro, A., Louckx, M. P., Vidal, H., et al. (1996). J. Biol. Chem. 271: 17875–17880.