Critical roles of conserved carboxylic acid residues in pigeon cytosolic NADP + -dependent malic enzyme Shuo-Chin Chang 1 *, Kuan-Yu Lin 1 *, Yu-Jung Chen 1 , Chin-Hung Lai 1 , Gu-Gang Chang 2 and Wei-Yuan Chou 1 1 Department of Biochemistry, National Defense Medical Center, Taipei, Taiwan 2 Faculty of Life Sciences, Institute of Biochemistry, Structural Biology Program, National Yang-Ming University, Taipei, Taiwan Cytosolic NADP + -dependent malic enzyme (EC 1.1.1.40) catalyses the decarboxylation of l-malate to pyruvate with oxaloacetate as intermediate and is asso- ciated with the reduction of NADP + to NADPH in the presence of a bivalent metal ion. Malic enzyme is a tetramer of identical subunits. The 3D structures of malic enzymes have been studied extensively [1]. Since the first crystal structure was solved for the human mitochondrial NAD(P) + -dependent malic enzyme complexed with Mn 2+ and ATP [2], 14 structures from various species, including the human, pigeon, the roundworm Ascaris suum, and the bacterium Thermo- toga maritima, have been deposited in the protein data- bank. All these structures, except that of T. maritima, have similar topology. These structures are classified into open and closed forms, depending on the presence of the substrate, l-malate, or its analogues [3]. It has been proposed that the closed form is the catalytically active form of the enzyme. Based on pH profiles and isotope studies of malic enzyme, it was proposed that its catalysis involves a general acid ⁄ base mechanism [4–7]. A general base is involved in deprotonating the C2 hydroxy group to form an oxaloacetate intermediate and in facilitating the hydride transfer from C2 to NADP + . After decarboxylation of oxaloacetate, a general acid partici- pates in the enol–keto tautomerization of pyruvate. Site-directed mutagenesis and kinetic results suggest that K199 in Ascaris (K162 in pigeon) [8] and D295 (D258 in pigeon) [9] function as the general acid and base, respectively. Our previous studies indicated that the K162 residue of pigeon NADP + -dependent malic Keywords chemical rescue; general acid ⁄ base; malic enzyme; metal ion binding; site-directed mutagenesis Correspondence W Y. Chou, Department of Biochemistry, National Defense Medical Center, 161 MinQuan E. Road Sec 6, Taipei, Taiwan 11490 Fax: +886 2 8792 3106 Tel: +886 2 8791 0776 E-mail: wyc@mail.ndmctsgh.edu.tw *These authors contributed equally to the experimental work. (Received 19 May 2006, revised 2 July 2006, accepted 7 July 2006) doi:10.1111/j.1742-4658.2006.05409.x Malic enzyme catalyses the reduction of NADP + to NADPH and the decarboxylation of l-malate to pyruvate through a general acid ⁄ base mech- anism. Previous kinetic and structural studies differ in their interpretation of the amino acids responsible for the general acid ⁄ base mechanism. To resolve this discrepancy, we used site-directed mutagenesis and kinetic ana- lysis to study four conserved carboxylic amino acids. With the D257A mutant, the K m for Mn 2+ and the k cat decreased relative to those of the wild-type by sevenfold and 28-fold, respectively. With the E234A mutant, the K m for Mg 2+ and l-malate increased relative to those of the wild-type by 87-fold and 49-fold, respectively, and the k cat remained unaltered, which suggests that the E234 residue plays a critical role in bivalent metal ion binding. The k cat for the D235A and D258A mutants decreased relative to that of the wild-type by 7800-fold and 5200-fold, respectively, for the over- all reaction, by 800-fold and 570-fold, respectively, for the pyruvate reduc- tion partial reaction, and by 371-fold and 151-fold, respectively, for the oxaloacetate decarboxylation. The activities of the overall reaction and the pyruvate reduction partial reaction of the D258A mutant were rescued by the presence of 50 mm sodium azide. In contrast, small free acids did not have a rescue effect on the activities of the E234A, D235A, and D257A mutants. These data suggest that D258 may act as a general base to extract the hydrogen of the C2 hydroxy group of l-malate with the aid of D235- chelated Mn 2+ to polarize the hydroxyl group. 4072 FEBS Journal 273 (2006) 4072–4081 ª 2006 The Authors Journal compilation ª 2006 FEBS enzyme is a general acid that donates a proton in enol–keto tautomerization [10]. However, the crystal structure of human mitochondrial NAD(P) + -depend- ent malic enzyme revealed that the oxygen of the carb- oxy group of D279 (D258 in the pigeon, D295 in A. suum) is structurally too distant to extract the pro- ton from the C2 hydroxy group of l-malate and would not play a role in the general acid ⁄ base mechanism [11]. Therefore, the authors proposed that K183 (K162 in the pigeon, K199 in A. suum) and Y112 (Y91 in the pigeon, Y126 in A. suum) are the general base and acid, respectively. Similar geometry was observed in A. suum mitochondrial NAD-dependent malic enzyme [12]. Recently, Cook and his colleagues [13] re-exam- ined the contribution of residues Y126, K199, and D294 (D257 in the pigeon, D274 in the human) to pH–rate profiles. They proposed that these three resi- dues form a catalytic triad, with K199 as the general base and Y126 as the general acid in the enzymatic mechanism. The crystal structure of pigeon malic enzyme showed that the metal ion is co-ordinated with the carboxylic group on the side chain of E234, D235, D258, the C1 carboxy group and the C2 hydroxy group of l-malate, and a free water molecule to form an octahedral con- formation [14]. The metal-binding roles of E234 and D235 have been confirmed in metal-protected urea- denaturation studies [15]. However, the K m value for Mn 2+ decreased 100-fold with E234Q, but was unal- tered with D235N. This prompted us to examine the contribution of these three amino-acid residues to metal ion binding and enzymatic catalysis. In this study, we sought to delineate the possible roles of these conserved residues in the active site of pigeon cytosolic NADP + -dependent malic enzyme by site-directed mutagenesis and detailed enzyme kinetic studies. Results Purification and structural characterization of wild-type and mutant malic enzymes To evaluate the possible roles of the conserved carb- oxylic amino acids at the active site, E234, D235, D257, and D258 in pigeon NADP + -dependent malic enzyme were replaced by alanine using site-directed mutagenesis. The mutated enzymes were expressed in Escherichia coli BL21(DE3) and purified. All recom- binant enzymes were shown to be homogeneous by SDS ⁄ PAGE (see Supplementary material Fig. S1). CD spectra of all recombinant enzymes were measured to evaluate whether the secondary structures of the mutant enzymes were altered. The CD spectra of the four mutant enzymes were very similar to that of the wild-type (Fig. S2). Differences in absorption intensity were caused by differences in protein concentration. All enzymes had similar contents of a-helix and b-sheet secondary structures. Most of the kinetic vari- ation in the mutant enzymes was caused by a lack of functional groups and not by global conformational changes. To determine whether malic enzyme endogenous to E. coli was present in our purified recombinant enzymes, an alternative construct of these mutants was expressed in the pET15b plasmid. These mutant enzymes, which contained a His 6 tag at the N-termi- nus, were purified using a Ni 2+ -chelating column to exclude endogenous malic enzyme. The enzymatic activities of these constructs were similar to those of enzymes that were purified using an ADP–Sepharose column (data not shown). This suggests that the amount of endogenous enzyme in our preparations was negligible. Steady-state kinetic properties of wild-type and mutant malic enzymes Preliminary kinetic studies showed that none of the mutants had an appreciable effect on the apparent K m for NADP + . Because the metal ion K m and k cat dif- fered between mutants, we performed detailed initial velocity studies in which both the metal ion and the l-malate concentrations were varied. The kinetic parameters of wild-type and mutant malic enzymes are summarized in Table 1. Replacement of residues D235 and D258 with alanine resulted in K m values for Mn 2+ similar to those of the wild-type. The k cat values of D235A and D258A were at least four orders of magni- tude less than that of the wild-type enzyme. These results suggest that the carboxy groups of D235 and D258 are essential for enzymatic catalysis. Of the three metal chelated amino-acid residues (E234, D235, and D258), only the E234A mutant demonstrated a sub- stantial decrease in affinity for bivalent metal ions. High concentrations of Mn 2+ resulted in the forma- tion of a brownish Mn–malate complex, which inter- fered with the enzyme assay. Therefore, Mg 2+ was used instead for kinetic studies of the E234A mutant. E234A had no effect on k cat , but induced 87-fold and 49-fold increases in K m values for Mg 2+ and l-malate, respectively. The D257A mutant had the least effect on the K m value for Mn 2+ (sevenfold decrease) and the k cat value (28-fold decrease), indicating that the D257 residue is not essential for metal ion binding and catalysis. The K m values for the metal ion and the S C. Chang et al. Mechanism of malic enzyme FEBS Journal 273 (2006) 4072–4081 ª 2006 The Authors Journal compilation ª 2006 FEBS 4073 substrate were similar to the corresponding K d values in the wild-type and in all mutants except the D257A mutant. This is in agreement with results showing that the release of NADPH is the rate-limiting step for pigeon NADP-dependent malic enzyme [16]. The D257A mutant caused a sevenfold decrease and a 4.5-fold increase in K m and K d values for the metal ion, respectively. This may be caused by perturbation of the network of hydrogen bonding in the D257A mutant [3]. Partial reactions catalysed by recombinant malic enzymes The reaction catalysed by malic enzymes consists of oxidoreduction and decarboxylation. The rate of each reaction can be measured independently of the other. The kinetics of these two reactions were examined in mutants that decreased k cat of the overall reaction (D235A, D257A, and D258A). The results of these studies are summarized in Table 2. The changes in the kinetic parameters of the D257A mutant were small relative to those of the D235A and D258A mutants (fourfold and twofold changes in k cat for oxidoreduc- tion and decarboxylation, respectively). However, the k cat values for both reactions changed substantially with both the D235A and D258A mutants. For the reduction of pyruvate (the reverse of oxidation of malate to oxaloacetate), the k cat values decreased 800-fold and 570-fold for D235A and D258A, respect- ively, and k cat values for the decarboxylation of oxalo- acetate decreased 371-fold and 151-fold for D235A and D258A, respectively. pH studies The pH–rate profile of wild-type enzyme showed a bell-shaped curve with pK a values of 6.29 ± 0.01 and 8.78 ± 0.09 at the acidic and basic sites, respectively. The pH–rate profiles for D235A, D257A, and D258A also showed bell-shaped curves, with two pK a values (Fig. 1). The estimated pK a values from the pH profile studies are summarized in Table 3. The acidic and basic pK a values for D258A were almost identical with those of the wild-type, and the differences were within the limits of experimental error. The acidic pK a values for D235A and D257A were also similar to that of the wild-type, but their basic pK a values were increased to 9.10 and 9.23, respectively. Chemical rescue experiments Amino-acid residues involved in general acid ⁄ base mechanisms can be identified using the chemical rescue method. The abilities of the sodium salts of formate, acetate, propionate, butanoate, and azide to rescue lost function of the E234A, D235A, D257A, and D258A mutants were studied. None of the small acids rescued the activities of mutants E234A, D235A, or D257A. The only restoration of activity occurred with the Table 1. Kinetic parameters for wild-type and mutant pigeon cytosolic NADP + -dependent malic enzymes. K mNADP (app) (l M) K mMal (mM) K dMal (mM) K mMn (lM) K dMn (lM) K mMg (mM) K dMg (mM) k cat (s )1 ) Wild-type (Mn 2+ ) 2.07 ± 0.15 0.08 ± 0.01 0.11 ± 0.02 3.78 ± 0.36 5.12 ± 0.87 31.34 ± 1.11 Wild-type (Mg 2+ ) 0.27 ± 0.03 0.27 ± 0.05 0.16 ± 0.02 0.15 ± 0.04 34.83 ± 2.32 E234A 1.80 ± 0.08 13.33 ± 3.26 13.06 ± 2.46 13.96 ± 1.82 13.09 ± 2.87 46.44 ± 5.81 D235A 1.79 ± 0.10 0.10 ± 0.01 0.19 ± 0.02 3.22 ± 0.14 6.55 ± 0.60 (0.04 ± 0.00) · 10 )1 D257A 2.83 ± 0.17 0.10 ± 0.01 0.65 ± 0.10 0.50 ± 0.07 23.36 ± 9.41 1.10 ± 0.04 D258A 2.96 ± 0.07 0.05 ± 0.01 0.15 ± 0.02 3.91 ± 0.24 12.69 ± 2.38 (0.06 ± 0.00) · 10 )1 Table 2. Kinetic parameters of partial reactions for wild-type and mutant malic enzymes. Reduction reaction Decarboxylation reaction K mPyr (app) (mM) k cat (app) (s )1 ) K mOAA (app) (mM) k cat (app) (s )1 ) Wild-type 6.05 ± 0.16 0.80 ± 0.01 0.17 ± 0.01 33.41 ± 0.58 D235A 6.11 ± 0.20 (0.10 ± 0.00) · 10 )2 0.91 ± 0.04 0.09 ± 0.00 D257A 1.42 ± 0.04 0.18 ± 0.0.00 0.07 ± 0.01 76.72 ± 2.02 D258A 1.90 ± 0.18 (0.14 ± 0.00) · 10 )2 2.37 ± 0.21 0.22 ± 0.01 Mechanism of malic enzyme S C. Chang et al. 4074 FEBS Journal 273 (2006) 4072–4081 ª 2006 The Authors Journal compilation ª 2006 FEBS D258A mutant in the presence of azide (Fig. 2A). Activation of D258A reached a maximum at 100 mm sodium azide and then declined at higher concentra- tions. The extent of activation was underestimated because of the presence of unsaturated l-malate and Mn 2+ in the assay mixture. To investigate the re-acti- vation process further, kinetic parameters for mutant D258A were determined in the presence of sodium azide (Table 4). Sodium azide had no significant effect on K m values for l-malate and Mn 2+ and on the k cat value when wild-type enzyme was used. With D258A, sodium azide increased the K m values for l-malate and Mn 2+ by 25-fold and 286-fold, respectively, compared with those observed in the absence of sodium azide. The k cat value for the D258A mutant was 890 times greater in the presence of the azide ion than in the absence of the azide ion (Tables 1 and 4). The activity of the D258A mutant was restored to 42% of that of the wild-type by sodium azide. To provide further insight into the catalytic roles of the D258 residue, the two partial reactions were examined by azide rescue. Only the pyruvate reduction reaction was rescued by sodium azide (Fig. 2B). The kinetic studies showed that the k cat value for D258A was identical with that Table 3. Summary of k cat pH data for wild-type and mutant malic enzymes. k cat (s )1 ) pK a1 pK a2 Wild-type 6.29 ± 0.01 8.78 ± 0.09 D235A 6.29 ± 0.09 9.10 ± 0.11 D257A 6.50 ± 0.04 9.23 ± 0.09 D258A 6.34 ± 0.08 8.72 ± 0.09 Fig. 1. pH–k cat profiles for wild-type and mutant pigeon cytosolic NADP + -dependent malic enzyme. The profiles for wild-type (s), D235A (n), D257A (m), and D258A (d) are shown. Malic enzyme activity was assayed as described in Experimental procedures. Points are the experimental data, and traces are the results of a fit of data for the pH–rate equation log y ¼ log[C ⁄ (1 + H ⁄ K a1 + K a2 ⁄ H)]. Fig. 2. Fold of activation of mutant malic enzyme as a function of the concentration of sodium azide. (A) The mutant malic enzyme overall oxidative decarboxylation activities of E234A (n), D235A (d), D257A (h), and D258A (s) were assayed as described in Experimental procedures. (B) The azide rescue of reduction partial reaction of wild-type (s) and D258A (d) and decarboxylation activ- ity of D258A (.). S C. Chang et al. Mechanism of malic enzyme FEBS Journal 273 (2006) 4072–4081 ª 2006 The Authors Journal compilation ª 2006 FEBS 4075 of the wild-type in the presence of azide (Table 4). The results of the chemical rescue studies suggest that D258 may act as a general base to extract the proton of the C2 hydroxy of l-malate to facilitate oxaloace- tate formation. Discussion In these studies, site-directed mutagenesis was used to evaluate the catalytic roles of four highly conserved acidic residues in the active site of pigeon NADP + - dependent malic enzyme. Steady-state kinetic charac- terization of the E234A, D235A, D257A, and D258A mutants suggests that the D257 residue is not directly involved in enzyme function. E234 is important for the binding of bivalent metal to the enzyme, and D235 and D258 play critical roles in catalysis. Our kinetic results for the pigeon D257A mutant differ from those for the corresponding mutant from A. suum. The K m and k cat values and the bell-shaped pH profile of the pigeon D257A mutant did not differ significantly from that of the wild-type enzyme. In con- trast, the corresponding mutant from A. suum, mutant D294A, had a k cat of about 13 000-fold less than that of the wild-type and exhibited a pH-independent pat- tern at the basic end of its pH range [13]. The A. suum mitochondrial enzyme is allosterically activated and inhibited by fumarate and ATP, respectively [17,18], whereas the pigeon cytosolic enzyme is not regulated by any known allosteric effector. The amino-acid sequences of these two isozymes show 55% identity and 73% similarity. Therefore, kinetic differences between pigeon and A. suum mutant enzymes are probably caused by differences in the microenviron- ments at their active sites. The 3D structure of pigeon malic enzyme showed that the metal ion was co-ordinated with the carboxy groups of the E234, D235, and D258 side chains, the carbonyl group of oxalate (an analogue of enolpyru- vate), and water to form an octahedral complex [14]. However, our kinetic studies show that only the E234A mutant has a significant effect on metal bind- ing. These results are consistent with previous studies in which the metal-binding ability of E234Q was decreased 100-fold, whereas D235N had little effect on the K m for Mn 2+ [20]. The unique kinetic properties of the E234A mutant probably result from the specific geometrical arrangement of E234. The carboxy groups of E234 and D235 and the C1 carboxy and C2 hydroxy groups of l-malate are coplanarly chelated with Mn 2+ D258 and water are located axially above and beneath this plane, respectively. In this plane, E234 and D235 are diagonally opposed to the C1 carboxy group of l-malate and the C2 hydroxy group of l-malate, respectively (Fig. 3). The interaction of Mn 2+ and the C1 carboxy group of l-malate should be strengthened by omitting the chelating of the carb- oxy group of the residue E234 at the opposite direction in E234A mutant. This trans effect will drive the Mn 2+ toward l-malate and therefore decrease the affinity of Mn 2+ for the carboxy groups of D235 and D258. This may account for the increase in K m when the E234 residue was mutated to alanine. The nominal change in K m values observed with the D235A and D258A mutant enzymes may reflect the elimination of an unfavourable repulsive interaction between the carboxy group and neighbouring negatively charged ligands. In previous Fe 2+ -ascorbate cleavage and site- directed mutagenesis studies, we proposed that D258 was involved in metal ion binding [19,20]. However, in those studies, of the four D258 mutants, only D258E Table 4. Kinetic parameters of overall and reduction partial reaction for wild-type and D258A mutant malic enzyme in the presence of 50 m M sodium azide. Oxidoreduction decarboxylation of malate reaction Reduction of pyruvate reaction K mMal (app) (mM) K mMn (app) (lM) k cat (app) (s )1 ) K mPyr (app) (mM) k cat (app) (s )1 ) Wild-type 0.18 ± 0.01 1.16 ± 0.05 27.80 ± 0.28 0.65 ± 0.04 0.87 ± 0.01 D258A 3.20 ± 0.20 (9.38 ± 0.62) · 10 2 11.76 ± 0.24 5.69 ± 0.74 0.84 ± 0.03 Fig. 3. Proposed mechanism for reduction step of pigeon cytosolic NADP + -dependent malic enzyme. The scheme is not meant to imply correct geometry or stereochemistry but simply to show the movement of protons and electrons. Mechanism of malic enzyme S C. Chang et al. 4076 FEBS Journal 273 (2006) 4072–4081 ª 2006 The Authors Journal compilation ª 2006 FEBS showed any measurable activity. Its K m value for Mn 2+ increased by  1600-fold. In the present study, larger amounts of enzyme were used and had a pro- nounced effect on the k cat value but no effect on metal ion affinity. The previously reported effects of D258E may have been caused by the extra methylene group, which would have perturbed the position of Mn 2+ rel- ative to the other amino-acid residues responsible for its binding. A significant decrease in the k cat value of the D235A mutant has not been reported previously. The carboxy group of D235, Mn 2+ , and the C2 hydroxy group of l-malate are linear with Mn 2+ at the centre. There- fore, it is impossible for the D235 residue to act as a general acid ⁄ base in the catalytic mechanism. Our chemical rescue and pH–rate profile results also sup- port this contention, which is based on crystal struc- ture. It has been proposed that the metal ion acts as a Lewis acid to stabilize the negatively charged transition state [21]. In D235A, because the interaction between the carboxy and Mn 2+ does not occur, the chelating ability of Mn 2+ for the C2 hydroxy group of l-malate is increased. This strong electron-withdrawing ability might propagate through the C2 hydroxy at the a-position to the C–H bond at the b-position and make the hydrogen atom partially positive. This effect might make hydride transfer impossible and inactivate the enzyme. In the wild-type enzyme, this metal- induced polarization will not extend to the b-position and will be limited to the C2 hydroxy group of l-ma- late. It will increase the acidity of the hydroxy group and facilitate the transfer of the proton from the hydroxy group to the general base residue and the hydride transfer to NADP + to complete the oxidore- duction reaction (Fig. 3). The metal ion will then inter- act with the carbonyl oxygen of oxaloacetate and facilitate the decarboxylation reaction to form enol- pyruvate [21]. Our kinetic data on the D235A mutant demonstrated a dramatic decrease in k cat values for the overall reaction and both partial reactions, which is in agreement with a model in which the metal ion partici- pates in both partial reactions. These results also indi- cate that both the Lewis acid metal ion and the general acid ⁄ base residue are important for the cata- lytic mechanism of malic enzyme. The D295 (D258 in the pigeon) residue in A. suum malic enzyme was identified as a general base by kin- etic and site-directed mutagenesis studies [9]. However, its role has been questioned because of the inaccessibil- ity of the carboxyl oxygen to the hydrogen of the hyd- roxy group of l-malate in human [3] and A. suum malic enzymes [11]. A similar topology was observed in pigeon NADP + -dependent malic enzymes, in which the distance between the carboxy oxygen of D258 and the C2 hydroxy is 3.47 A ˚ . However, our kinetic studies showed that substitution of alanine for aspartate at the D258 residue decreased k cat values in the overall oxida- tive decarboxylation reaction and in the pyruvate reduction partial reaction. Both these enzymatic activ- ities of D258A mutant could be rescued by sodium azide. No azide rescue was observed for the decarb- oxylation partial reaction. These results indicate that the carboxylic group of D258 is essential for the first step of the enzymatic reaction in which a general base is involved. Sodium azide rescue has been widely used to distinguish nucleophile residues from general bases in glycosidases, in which azide can act as nucleophile but not as a proton acceptor [22]. However, the azide ion was shown to act as an exogenous proton acceptor in the re-activation of the acid ⁄ base mutants of Ther- mobacillus xylanilyticus a-l-arabinofuranosidase [23] and human b-glucuronidase [24]. Therefore, despite the contradiction between crystal structure and kinetic studies, we suggest that D258 might still act as a gen- eral base to accept a proton from the C2 hydroxy group to form a ketone and facilitate C2 hydride transfer (Fig. 3). The pH dependence of k cat has been interpreted as ionization of an enzymatic carboxy group essential for catalysis. The unexpected bell-shaped pH profile of the D258A mutant indicated that the acidic pK a may derive from chemical components other than the carb- oxy group of the D258 residue. The conditions used in the current studies were not acidic enough to reveal the pK a of the carboxy group of l-malate. Recently, studies showed that the pK a of the deprotonation of the metal-co-ordinated hydroxy group of isocitrate in the porcine mitochondrial NADP + -dependent isoci- trate dehydrogenase could be shifted to pH 5 [25]. Therefore, deprotonation of the metal-chelated hyd- roxy group substrate l-malate may be another reason for the acidic pK a in the pH profile. There are several possible reasons for the discrep- ancy between the results of studies of kinetics and those of studies of crystal structure. Firstly, the D235 and D258 mutants had the most profound effect on k cat values. This suggests that polarization of the C2 hydroxy group and the general acid ⁄ base reaction co-operatively extract the hydrogen of the C2 hydroxy group to facilitate hydride transfer. Therefore, the carboxylic group of D258, a weak base because it is relatively distal to the hydroxy group of l-malate, may still be able to act as a general base for the oxidore- duction reaction. Secondly, an active-site water mole- cule may exist between the carboxy group of D258 and the hydroxy group of l-malate and serve as a S C. Chang et al. Mechanism of malic enzyme FEBS Journal 273 (2006) 4072–4081 ª 2006 The Authors Journal compilation ª 2006 FEBS 4077 proton relay to fulfil the general base role of D258. Similar active-site water molecules have been observed in the crystal structure of porcine mitochondrial isoci- trate dehydrogenase, another oxidoreductive decarbox- ylated enzyme [26]. In this case, an aspartate residue and two water molecules form a catalytic triad that is responsible for the general base mechanism. Finally, the crystal structures of malic enzyme were solved in either the E–NADH–malate–Mn 2+ –fumarate penten- ary complex (human) or the E–NAD(P)H–oxalate– Mn 2+ tertiary complex (pigeon and Ascarid). Pigeon and Ascarid malic enzymes show substrate inhibition in the presence of a high concentration of l-malate [21,27,28]. It has been suggested that the substrate inhibition might result from the formation of an E–malate–NADPH–Mn 2+ aborted complex. Early kinetic studies showed that oxalate, an analogue of enolpyruvate, is a dead-end inhibitor for malic enzyme [29]. Therefore, all the 3D structures of the malic enzyme examined might represent inactive aborted enzymatic forms. The inhibition observed in the kinetic studies may have resulted from inaccessibility between the carboxy group of D258 and the C2 hydroxy group of l-malate. Therefore, the carboxy group of D258 may still be close enough to act as a general base to extract the C2 hydroxy proton of l-malate in the enzy- matically active complex. In conclusion, we have described the functional roles of these conserved carboxylic acid amino-acid residues using site-directed mutagenesis and steady-state kinet- ics. We propose the following: l E234 is essential for Mn 2+ binding. l The carboxy groups of D235 and D258 act co-operatively. l The D235 residue is involved in the polarization of the hydroxy group of l-malate by chelating the Mn 2+ ion. l The D258 residue acts as a general base to promote oxaloacetate formation and hydride transfer. Experimental procedures Materials Restriction endonucleases, T4 DNA polymerase, T4 DNA ligase, and T4 polynucleotide kinase were purchased from Promega (Madison, WI, USA). Q Sepharose and 2¢,5¢- ADP–Sepharose were obtained from Amersham (Piscata- way, NJ, USA). The pET21b expression vector was purchased from Novagen (Madison, WI, USA). NADP + was purchased from Sigma (St Louis, MO, USA). All other reagents were of molecular biology grade or the highest grade available. Cloning of pigeon liver malic enzyme cDNA The full-length pigeon liver cytosolic malic enzyme cDNA was cloned into the pET21b vector for expression, as previ- ously described [30]. The construction was designed in such a way that no extra nucleotide sequence flanked the 5¢ end of the ORF of the malic enzyme cDNA. Therefore, the amino-acid composition and sequence of the recombinant form were identical with those of the native enzyme. The plasmid containing malic enzyme cDNA was named pET21-ME. Site-directed mutagenesis Site-directed mutagenesis was carried out according to the procedures of Zoller & Smith [31] using the M13 origin in the vector for uracil-containing ssDNA preparation. Other DNA techniques were performed according to the protocols of Sambrook et al. [32]. The pET21-ME recombinant phagemid was amplified in the ung – and dut – CJ236 E. coli strain with helper phage R408 for preparation of the uracil- containing ssDNA template. The uracil-containing template DNA was annealed with phosphorylated mutagenic oligo- nucleotides and then extended in vitro and ligated by T4 DNA polymerase and T4 DNA ligase, respectively. The mutated DNA was screened by transforming into the ung + and dut + JM109 E. coli strain, and the surviving colonies were further identified by dideoxy chain-termination sequencing [33]. The entire cDNA was also sequenced to exclude any unexpected mutations resulting from in vitro DNA polymerase extension. Expression and purification of recombinant malic enzymes Expression plasmids for wild-type malic enzyme and mutants were introduced into the host E. coli BL21(DE3) and grown in Luria–Bertani medium containing 0.1 mgÆmL )1 ampicillin at 37 °CtoanA 660 of 0.5–0.6. Expression was induced with 1.0 mm isopropyl b-d-thiogalactopyranoside. The culture was then allowed to grow overnight at 25 °C. The cells were harvested by centrifugation for 15 min at 5000 g. Cells were resuspended and sonicated in Tris ⁄ HCl buffer (25 mm, pH 7.5) containing 2 mm 2-mercaptoethanol. The recombinant proteins were purified using a Q-Sepharose column pre-equilibrated with the same buffer. Malic enzyme was eluted with Tris ⁄ HCl buffer containing 150 mm NaCl. The fractions containing malic enzyme were further purified using a 2¢,5¢-ADP–Sepharose column. The malic enzyme was then eluted by 230 lm NADP + .A Sephadex G-25 gel filtration column was used to remove NADP + . All purified enzymes were subjected to SDS ⁄ PAGE to examine their purity. Protein concentrations were determined by the Bradford method using BSA as a standard [34]. Mechanism of malic enzyme S C. Chang et al. 4078 FEBS Journal 273 (2006) 4072–4081 ª 2006 The Authors Journal compilation ª 2006 FEBS CD measurements CD measurements were made with a Jasco J-810 spectropo- larimeter using a 0.1-cm path-length cell and averaging five repeated scans between 250 and 200 nm. Typically, 30 lg of the wild-type or mutated NADP + -dependent malic enzyme in Tris ⁄ HCl buffer (25 mm, pH 7.5) containing 2mm 2-mercaptoethanol was used for each measurement. The spectra were analysed on DICHROWEB (http:// www.cryst.bbk.ac.uk/cdweb/html/home.html) using the software of CDSSTR [35,36]. Enzyme assay Malic enzyme activity was assayed as described by Hsu & Lardy [37]. The reaction mixture contained triethanol- amine ⁄ HCl buffer (66.7 mm, pH 7.4), l-malate (5 mm), NADP + (0.23 mm), Mn 2+ (4 mm), and an appropriate amount of enzyme in a total volume of 1 mL. The forma- tion of NADPH at 25 °C was monitored continuously at 340 nm with a Perkin–Elmer Lambda 3B spectrophoto- meter. One unit of enzyme activity was defined as the initial rate of 1 lmol NADPH formed per minute under the assay conditions. A molar absorption coefficient of 6.22 · 10 3 m )1 Æcm )1 for NADPH was used in the cal- culations. Specific activity was defined as lmol NADPH formedÆmin )1 Æ(mg protein) )1 . Kinetic analysis Apparent Michaelis constants for the substrates were determined by varying one substrate concentration around its K m value while maintaining the other components con- stant. Initial velocity studies were performed to determine the Michaelis and dissociation constants for l-malate and Mn 2+ . For initial velocity studies, the concentrations of both l-malate and Mn 2+ were varied while that of NADP + was maintained at saturation. The E234A mutant required a higher concentration of Mn 2+ for initial velo- city studies than the other mutants. Under these condi- tions, a brownish Mn–malate complex formed, which would have interfered with the enzyme assay. Therefore, Mn 2+ was replaced by Mg 2+ for initial velocity studies of the E234A mutant. Concentrations of the other compo- nents were held constant. Data were analysed using the following equation, which describes a sequential initial velocity pattern: t ¼ V max AB=ðK ia K b þ K a B þ K b A þ ABÞ in which t and V max represent initial and maximum veloci- ties, A and B represent reactant concentrations, K a and K b are Michaelis constants for A and B, and K ia is the dissoci- ation constant for A. The linear regression analysis was carried out with commercial pro fit 6.0 (QuantumSoft, Uetikon am See, Switzerland). Partial reaction analysis The two partial activities of malic enzyme, decarboxylation and reduction, can be evaluated separately. The decarboxy- lation activity of malic enzyme was assayed by the method of Tang & Hsu [38] using oxaloacetate as substrate. The rate of decarboxylation of oxaloacetate was measured by monitoring the disappearance of the enolic oxaloacetate absorbance at 260 nm in the presence of Mn 2+ or Mg 2+ . Various concentrations of oxaloacetate in 185 mm potas- sium acetate buffer, pH 4.5, were added to 50 mm EDTA and incubated at 25 °C for 10 min to reach keto–enol equi- librium. The oxaloacetate solutions were added to a total volume of 1 mL containing 4 mm MnCl 2 and 37 mm potas- sium acetate buffer, pH 4.5 to start the reaction. The rate of decarboxylation in the presence of enzyme was corrected by subtracting the spontaneous oxaloacetate decarboxyla- tion. Oxidation of l-malate to oxaloacetate cannot be evalu- ated directly because of interference by the subsequent decarboxylation. The reversed direction, reduction of a-oxo acid to a-hydroxy acid, can be analysed using pyruvate and NADPH as substrates. The reduction partial reaction was performed as described by Tang & Hsu [39] using pyruvate as substrate. The rate of reduction of pyruvate to lactate was measured at 25 °C by monitoring the decrease in absorbance at 340 mm associated with the oxidation of NADPH. A typical assay mixture contained 66.7 mm tri- ethanolamine ⁄ HCl buffer (pH 7.4), 0.23 mm NADPH, 4mm MnCl 2 , 1–50 m m pyruvate (pH 7.4), and an appro- priate amount of malic enzyme. pH studies The pH dependencies of k cat for wild-type and mutants were determined using initial velocity studies and variable concentrations of l-malate and NADP + as a function of pH over the pH range 5.5–10, which was maintained with 60 mm Bis-Tris propane buffer. The pH values were recor- ded and showed no significant change before and after the initial velocity was measured. The pK a values were obtained by fitting the following equation to the data: log y ¼ log½C=ð1 þ H=K a1 þ K a2 =HÞ where y is the value of the parameter of interest (k cat ), C is the pH-independent value of y, H is the hydrogen ion concentration, and K a1 and K a2 are the acid dissociation constants for functional groups in the enzyme–substrate complex. Chemical rescue The stock solutions of exogenous acids were prepared at pH 7.4. Various free acids, including formic acid, acetic acid, butyric acid, or sodium azide, were added to the S C. Chang et al. Mechanism of malic enzyme FEBS Journal 273 (2006) 4072–4081 ª 2006 The Authors Journal compilation ª 2006 FEBS 4079 standard reaction mixture to examine their rescue abilities. To measure the kinetic properties of malic enzyme after rescue, 50 mm sodium azide was included for all kinetic studies. Acknowledgements This research was supported by a grant from the National Science Council, China (NSC92-2320-B016- 060 to W.Y.C.). We thank Dr Chi-Ching Hwang (Kaohsiung Medical University, Taiwan) and Dr Minghuey Shieh (National Taiwan Normal University, Taiwan) for helpful discussions. References 1 Chang GG & Tong L (2003) Structure and function of malic enzymes, a new class of oxidative decarboxylases. Biochemistry 42, 12721–12733. 2 Xu Y, Bhargava G, Wu H, Loeber G & Tong L (1999) Crystal structure of human mitochondrial NAD(P) + -dependent malic enzyme: a new class of oxidative decarboxylases. Struct Fold Des 7, R877–R889. 3 Yang Z, Floyd DL, Loeber G & Tong L (2000) Struc- ture of a closed form of human malic enzyme and impli- cations for catalytic mechanism. Nat Struct Biol 7, 251– 257. 4 Kiick DM, Harris BG & Cook PF (1986) Protonation mechanism and location of rate-determining steps for the Ascaris suum nicotinamide adenine dinucleotide- malic enzyme reaction from isotope effects and pH stu- dies. Biochemistry 25, 227–236. 5 Park SH, Harris BG & Cook PF (1986) pH dependence of kinetic parameters for oxalacetate decarboxylation and pyruvate reduction reactions catalyzed by malic enzyme. Biochemistry 25, 3752–3759. 6 Weiss PM, Gavva SR, Harris BG, Urbauer JL, Cleland WW & Cook PF (1991) Multiple isotope effects with alternative dinucleotide substrates as a probe of the malic enzyme reaction. Biochemistry 30, 5755–5763. 7 Karsten WE & Cook PF (1994) Stepwise versus con- certed oxidative decarboxylation catalyzed by malic enzyme: a reinvestigation. Biochemistry 33, 2096–2103. 8 Liu D, Karsten WE & Cook PF (2000) Lysine 199 is the general acid in the NAD-malic enzyme reaction. Biochemistry 39, 11955–11960. 9 Karsten WE, Chooback L, Liu D, Hwang CC, Lynch C & Cook PF (1999) Mapping the active site topo- graphy of the NAD-malic enzyme via alanine-scanning site-directed mutagenesis. Biochemistry 38, 10527– 10532. 10 Kuo CC, Tsai LC, Chin TY, Chang GG & Chou WY (2000) Lysine residues 162 and 340 are involved in the catalysis and coenzyme binding of NADP( + )-dependent malic enzyme from pigeon. Biochem Biophys Res Commun 270, 821–825. 11 Tao X, Yang Z & Tong L (2003) Crystal structures of substrate complexes of malic enzyme and insights into the catalytic mechanism. Structure (Camb) 11, 1141– 1150. 12 Rao GS, Coleman DE, Karsten WE, Cook PF & Harris BG (2003) Crystallographic studies on Ascaris suum NAD-malic enzyme bound to reduced cofactor and identification of an effector site. J Biol Chem 278, 38051–38058. 13 Karsten WE, Liu D, Rao GS, Harris BG & Cook PF (2005) A catalytic triad is responsible for acid-base chemistry in the Ascaris suum NAD-malic enzyme. Biochemistry 44 , 3626–3635. 14 Yang Z, Zhang H, Hung HC, Kuo CC, Tsai LC, Yuan HS, Chou WY, Chang GG & Tong L (2002) Structural studies of the pigeon cytosolic NADP( + )-dependent malic enzyme. Protein Sci 11, 332–341. 15 Chang HC, Chou WY & Chang GG (2002) Effect of metal binding on the structural stability of pigeon liver malic enzyme. J Biol Chem 277, 4663–4671. 16 Schimerlik MI, Grimshaw CE & Cleland WW (1977) Determination of the rate-limiting steps for malic enzyme by the use of isotope effects and other kinetic studies. Biochemistry 16, 571–576. 17 Lai CJ, Harris BG & Cook PF (1992) Mechanism of activation of the NAD-malic enzyme from Ascaris suum by fumarate. Arch Biochem Biophys 299, 214–219. 18 Landsperger WJ & Harris BG (1976) NAD + -malic enzyme. Regulatory properties of the enzyme from Ascaris suum. J Biol Chem 251, 3599–3602. 19 Wei CH, Chou WY, Huang SM, Lin CC & Chang GG (1994) Affinity cleavage at the putative metal-binding site of pigeon liver malic enzyme by the Fe( 2+ )-ascor- bate system. Biochemistry 33, 7931–7936. 20 Wei CH, Chou WY & Chang GG (1995) Identification of Asp258 as the metal coordinate of pigeon liver malic enzyme by site-specific mutagenesis. Biochemistry 34, 7949–7954. 21 Hsu RY, Mildvan AS, Chang G & Fung C (1976) Mechanism of malic enzyme from pigeon liver. Mag- netic resonance and kinetic studies of the role of Mn 2+ . J Biol Chem 251, 6574–6583. 22 Zechel DL & Withers SG (2001) Dissection of nucleo- philic and acid-base catalysis in glycosidases. Curr Opin Chem Biol 5, 643–649. 23 Debeche T, Bliard C, Debeire P & O’Donohue MJ (2002) Probing the catalytically essential residues of the alpha-L-arabinofuranosidase from Thermobacillus xylan- ilyticus. Protein Eng 15, 21–28. 24 Islam MR, Tomatsu S, Shah GN, Grubb JH, Jain S & Sly WS (1999) Active site residues of human beta-glu- curonidase. Evidence for Glu (540) as the nucleophile Mechanism of malic enzyme S C. Chang et al. 4080 FEBS Journal 273 (2006) 4072–4081 ª 2006 The Authors Journal compilation ª 2006 FEBS and Glu (451) as the acid-base residue. J Biol Chem 274, 23451–23455. 25 Huang YC, Grodsky NB, Kim TK & Colman RF (2004) Ligands of the Mn 2+ bound to porcine mitoch- ondrial NADP-dependent isocitrate dehydrogenase, as assessed by mutagenesis. Biochemistry 43, 2821–2828. 26 Ceccarelli C, Grodsky NB, Ariyaratne N, Colman RF & Bahnson BJ (2002) Crystal structure of porcine mito- chondrial NADP + -dependent isocitrate dehydrogenase complexed with Mn 2+ and isocitrate. Insights into the enzyme mechanism. J Biol Chem 277, 43454–43462. 27 Park SH, Harris BG & Cook PF (1989) Substrate acti- vation by malate induced by oxalate in the Ascaris suum NAD-malic enzyme reaction. Biochemistry 28, 6334– 6340. 28 Pry TA & Hsu RY (1980) Equilibrium substrate binding studies of the malic enzyme of pigeon liver. Equivalence of nucleotide sites and anticooperativity associated with the binding of l-malate to the enzyme-manganese(II)- reduced nicotinamide adenine dinucleotide phosphate ternary complex. Biochemistry 19, 951–962. 29 Hsu RY (1982) Pigeon liver malic enzyme. Mol Cell Biochem 43, 3–26. 30 Chou WY, Huang SM, Liu YH & Chang GG (1994) Cloning and expression of pigeon liver cytosolic NADP( + )-dependent malic enzyme cDNA and some of its abortive mutants. Arch Biochem Biophys 310, 158–166. 31 Zoller MJ & Smith M (1982) Oligonucleotide-directed mutagenesis using M13-derived vectors: an efficient and general procedure for the production of point mutations in any fragment of DNA. Nucleic Acids Res 10, 6487– 6500. 32 Sambrook J, Fritsch EF & Maniatis T (1989) Molecular Cloning: a Laboratory Manual. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY. 33 Sanger F, Nicklen S & Coulson AR (1977) DNA sequencing with chain-terminating inhibitors. Proc Natl Acad Sci USA 74 , 5463–5467. 34 Bradford MM (1976) A rapid and sensitive method for the quantitation of microgram quantities of protein util- izing the principle of protein-dye binding. Anal Biochem 72, 248–254. 35 Whitmore L & Wallace BA (2004) DICHROWEB, an online server for protein secondary structure analyses from circular dichroism spectroscopic data. Nucleic Acids Res 32, W668–W673. 36 Sreerama N & Woody RW (2000) Estimation of protein secondary structure from circular dichroism spectra: comparison of CONTIN, SELCON, and CDSSTR methods with an expanded reference set. Anal Biochem 287, 252–260. 37 Hsu RY & Lardy HA (1967) Pigeon liver malic enzyme. II. Isolation, crystallization, and some properties. J Biol Chem 242, 520–526. 38 Tang CL & Hsu RY (1974) Mechanism of pigeon liver malic enzyme. Modification of sulfhydryl groups by 5,5¢-dithiobis(2-nitrobenzoic acid) and N-ethylmalei- mide. J Biol Chem 249, 3916–3922. 39 Tang CL & Hsu RY (1973) Reduction of alpha-oxo carboxylic acids by pigeon liver ‘malic’ enzyme. Biochem J 135, 287–291. Supplementary material The following supplementary material is available online: Fig. S1. SDS-PAGE of purified wild-type and mutant malic enzymes. Fig. S2. CD spectra of wild-type and mutated pigeon NADP-malic enzyme. This material is available as part of the online article from http://www.blackwell-synergy.com S C. Chang et al. Mechanism of malic enzyme FEBS Journal 273 (2006) 4072–4081 ª 2006 The Authors Journal compilation ª 2006 FEBS 4081 . Critical roles of conserved carboxylic acid residues in pigeon cytosolic NADP + -dependent malic enzyme Shuo-Chin Chang 1 *, Kuan-Yu Lin 1 *, Yu-Jung. catalytic roles of four highly conserved acidic residues in the active site of pigeon NADP + - dependent malic enzyme. Steady-state kinetic charac- terization of