Alteration of Substrate Specificity of Leucine Dehydrogenase by

Title Alteration of substrate specificity of leucine dehydrogenase by site-directed mutagenesis Author(s) 片岡, 邦重; Kataoka, Kunishige; Tanizawa, Katsuyuki Citation Journal of molecular catalysis b enzymatic, 23(2-6): 299-309 Issue Date 2003-09 Type Journal Article Text version author URL http://hdl.handle.net/2297/1731 Right *KURAに登録されているコンテンツの著作権は，執筆者，出版社（学協会）などが有します。 *KURAに登録されているコンテンツの利用については，著作権法に規定されている私的使用や引用などの範囲内で行ってください。 *著作権法に規定されている私的使用や引用などの範囲を超える利用を行う場合には，著作権者の許諾を得てください。ただし，著作権者 から著作権等管理事業者（学術著作権協会，日本著作出版権管理システムなど）に権利委託されているコンテンツの利用手続については ，各著作権等管理事業者に確認してください。 http://dspace.lib.kanazawa-u.ac.jp/dspace/ Journal of Molecular Catalysis B: Enzymatic (Regular paper) Title: Alteration of Substrate Specificity of Leucine Dehydrogenase by Site-directed Mutagenesis1 Authors: Kunishige Kataokaa,* and Katsuyuki Tanizawab Affiliations: a Department of Chemistry, Faculty of Science, Kanazawa University, Kakuma, Kanazawa, Ishikawa 920-1192, Japan b Department of Structural Molecular Biology, Institute of Scientific and Industrial Research, Osaka University, Ibaraki, Osaka 567-0047, Japan *Correspondence auther: (kataoka@cacheibm.s.kanazawa-u.ac.jp). 1 Dedicated to Professor Dr. Kenji Soda in honor of his 70th birthday. 1 Abstract The residues L40, A113, V291, and V294, in leucine dehydrogenase (LeuDH), predicted to be involved in recognition of the substrate side chain, have been mutated on the basis of the molecular modeling to mimic the substrate specificities of phenylalanine (PheDH), glutamate (GluDH), and lysine dehydrogenases (LysDH). The A113G and A113G/V291L mutants, imitating the PheDH active site, displayed activities toward L-phenylalanine and phenylpyruvate with 1.6 and 7.8% of kcat values of the wild-type enzyme for the preferred substrates, L-leucine and its keto-analog, respectively. Indeed, the residue A113, corresponding to G114 in PheDH, affects the volume of the side-chain binding pocket and has a critical role in discrimination of the bulkiness of the side chain. Another two sets of mutants, substituting L40 and V294 of LeuDH with the corresponding residues predicted in GluDH and LysDH, were also constructed and characterized. Emergence of GluDH and LysDH activities in L40K/V294S and L40D/V294S mutants, respectively, indicates that the two corresponding residues in the active site of amino acid dehydrogenases are important for discrimination of the hydrophobicity/polarity of the aliphatic substrate side chain. All these results demonstrate that the substrate specificities of the amino acid dehydrogenases can be altered by protein engineering. The engineered dehydrogenases are expected to be used for production and detection of natural and non-natural amino acids. Key words: leucine dehydrogenase, substrate specificity, protein engineering, enzymatic synthesis 2 1. Introduction Leucine dehydrogenase (LeuDH) [EC 1.4.1.9] is an NAD+- dependent oxidoreductase that catalyzes the reversible deamination of L-leucine and some other branched-chain L-amino acids to their keto analogs (Scheme 1). The enzyme occurs ubiquitously in Bacillus species [1] and functions catabolically in the bacterial metabolism of branched-chain L-amino acids [2]. It has been suggested that the enzyme plays an important role in spore germination in cooperation with alanine dehydrogenase [3, 4]. The amino acid dehydrogenase has considerable commercial potential for the production of novel nonproteinous amino acids in pharmaceutical industries [5, 6] and for the diagnosis of genetic diseases of amino acid metabolisms including phenylketonuria [7], maple syrup urine disease [8, 9], and homocystinuria [10]. H2 N + + O + NAD + H2O! COOH + + NH 4 + NADH + H COOH Scheme 1 The thermostable LeuDH cloned from Bacillus stearothermophilus [11] shares considerable sequence similarities in the catalytic and coenzyme-binding domains with the enzymes acting on other amino acids, such as glutamate (GluDH) [12, 13], phenylalanine (PheDH) [14, 15], and valine (ValDH) [16, 17] dehydrogenases. Although the overall similarities among these enzymes are not high, sequence similarities between these dehydrogenases clearly indicate the existence of an enzyme superfamily related by divergent evolution [18]. However, the substrate specificities of this superfamily are different; GluDH 3 recognizes and binds glutamate in preference to all other amino acids [19], LeuDH and ValDH catalyze the oxidation of only branched-chain, aliphatic amino acids [1, 20, 21], and PheDH has a marked preference for aromatic amino acids as its substrate, although it also accepts smaller hydrophobic amino acids with reduced efficiency [22, 23]. Results of structural studies in the last decade have revealed that these enzymes share a very similar subunit structure, with each subunit composed of two domains separated by a cleft harboring the active site [24-26]. To study the structure and function of the domains of amino acid dehydrogenases before their crystal structures were available, we constructed and characterized the chimeric enzyme consisting of an N-terminal domain of thermostable PheDH from Thermoactinomyces intermedius and a C-terminal domain of LeuDH from B. stearothermophilus [27]. Furthermore, we constructed and expressed genes of fragmentary forms of B. stearothermophilus LeuDH to investigate the function of the enzyme domains [28]. The results of our domain manipulation have suggested that the substrate specificity is determined by structural interactions of the two domains, as corroborated from the presently known crystal structures. On the basis of the molecular structures of GluDH [24], LeuDH [25], and PheDH [26], substrate specificity of amino acid dehydrogenases is explained by the types of residues comprising the substrate side-chain binding pocket. In GluDH of Clostridium symbiosum, the principal interactions that determine the specificity are between the γ-carboxyl group of the substrate glutamate and the amino group of K89 and the hydroxyl group of S380, with G90, A163, and V377 interacting with the hydrophobic component of the glutamate side chain [29]. In the Bacillus sphaericus LeuDH structure, the latter three residues are conserved (G41, A113, and V291, respectively), whereas K89 and S380 in GluDH are replaced by L40 and V294 in LeuDH (residue numbers refer to B. sphaericus LeuDH), making up a more 4 hydrophobic substrate side-chain binding pocket [25]. X-ray crystallographic and homologybased molecular modeling studies of PheDH revealed that this enzyme has a hydrophobic and large side-chain binding pocket enough to bind the phenylalanine benzene ring by replacing glycine for A113 of LeuDH [18, 30]. Several attempts have been made to alter the substrate specificities of the amino acid dehydrogenase superfamily by molecular modeling. Wang et al. engineered a K89L/S380V double mutant of GluDH, expecting that it would show substrate specificity of LeuDH with a hydrophobic pocket [30], but the mutant had no activity because of steric clash of the pocket [31]. The substrate specificity of B. sphaericus PheDH was successfully shifted toward that of LeuDH by replacing G124 and L307 (corresponding to A163 and V377 in GluDH, respectively) to alanine and valine, respectively [32]. More recently, Seah et al. succeeded in modulating the specificity of B. sphaericus PheDH [33]. In the present study, aiming at altering the substrate specificity of B. stearothermophilus LeuDH by homology-based modeling, we chose L40, A113, V291, and V294 for a target of mutagenesis to mimic the substrate-binding pocket of PheDH, GluDH, and LysDH. The detailed kinetic analysis of the mutant enzymes suggests that these residues are important in the substrate side-chain recognition of the enzyme superfamily. The alteration of substrate specificity by protein engineering presented here is not only of scientific interest – understanding the structural basis for the difference in substrate discrimination of the amino acid dehydrogenases – but also of applied one, possibly providing an amino acid dehydrogenase with novel properties for production and detection of natural and non-natural amino acids. 5 2. Experimental 2.1. Materials A site-directed mutagenesis kit (Mutan-K), Taq DNA polymerase, T4 DNA ligation kit, and DNA blunting kit were purchased from Takara Biochemicals (Japan). All reagents for DNA synthesis and DNA sequencing were purchased from Applied Biosystems (U.S.A.). The plasmid pICD2 carrying the leucine dehydrogenase gene of B. stearothermophilus and the plasmid pKPDH2 containing the phenylalanine dehydrogenase gene of T. intermedius were described previously [11, 14]. 2.2. Mutagenesis Substitutions of aspartate or lysine for L40, glycine for A113, leucine for V291, and serine for V294 of LeuDH were performed by the method of Kunkel et al. [34] using a commercial kit (Mutan-K), as described previously [35]. The following five oligonucleotide primers were synthesized with an Applied Biosystems DNA synthesizer model 381 to contain appropriate mismatched bases (indicated by asterisks) in the complementary codons for each residue (underlined):         *** L40D: 5’-ATACGCGTCCCGCCTTTCGCCGGGCCGAG-3’         ** L40K: 5’-ATACGCGTCCCGCCGTCCGCCGGGCCGAG-3’        * A113G: 5’-GACGTCTTCCCCCGTGATGTAG-3’        * V291L: 5’-ACGTTGATGAGGCCGCCGGC-3’ ** V294S: 5’-TTCGTCCGCGGAGGTTGATGACG-3’ 6 After confirming the nucleotide sequence of the mutant gene, appropriate restriction fragments were replaced for the corresponding wild-type gene fragments in the LeuDH expression plasmid pICD212. Substitution of alanine for G114 of PheDH was performed by the two-step PCR method as described previously [36] with two mutation primers shown below and pKPDH2 as a template (1st PCR): * G114A-sense: 5’-AACGGCCGTTTCTATACCGCAACCGACATGGG-3’ * G114A-antisense: 5’-CCATGTCGGTTGCGGTATAGAAAC-3’ After the second PCR using universal primers, the amplified fragment was sequenced and digested with SacII and SplI. The digested fragment was replaced for the corresponding wild-type gene fragments in pKPDH2. 2.3. Purification of the mutant enzyme The wild-type and mutant enzymes of LeuDH were purified to homogeneity from the crude extract of recombinant Escherichia coli cells grown at 37 °C for 12 h in Luria broth supplemented with 50 µg/ml ampicillin and 0.5 mM IPTG, as described previously [11]. The wild-type and G114A mutant of PheDH were also purified to homogeneity from the crude extract of recombinant E. coli cells, as described previously [14]. 2.4. Steady-state kinetic analysis The oxidative deamination of L-amino acids and the reductive amination of α-keto acids were measured by monitoring spectrophotometrically the appearance and disappearance, 7 respectively, of NADH under the conditions described previously [11]. The steady-state kinetic parameters were determined by varying systematically the concentrations of both substrate and coenzyme, except for ammonia, which was held at a constant, saturating concentration (1.0 M). Protein concentrations of the wild-type and mutant LeuDH and the wild-type and G114A PheDH were estimated using absorbencies (A0.1%) of 0.851 and 0.634 at 280 nm, respectively. 2.5. CD measurements Circular dichroism (CD) spectra were measured at 25 °C in 10 mM potassium phosphate buffer (pH 7.2) with a Jasco spectropolarimeter model J-600. In the calculation of the mean residue ellipticity (s), the mean residue weight was taken to be 111 for the enzyme protein. The CD spectra were obtained at a protein concentration of 0.2 mg/ml in a 2.0-cm light path length cuvette for the measurements in the wavelength region above 250 nm and at a protein concentration of 0.1 mg/ml in a 0.1-cm light path length cuvette for measurements below 250 nm. 8 3. Results and Discussion 3.1. Molecular modeling and construction of mutant enzymes Multiple sequence alignment of LeuDH, PheDH, and GluDH using structural information of B. sphaericus LeuDH and C. symbiosum GluDH pointed out important residues that are assumed to be the determinant of substrate side-chain binding (Fig. 1). This homology-based modeling suggests that the difference in substrate specificity between PheDH and LeuDH arises only from unacceptable steric interaction of the methyl group of A113 of LeuDH with the substrate phenylalanine benzene ring, which are relieved in PheDH by critical replacement of this residue by Gly114 [18]. To examine the validity of the models, two mutant enzymes of LeuDH, in which Ala113 is replaced by glycine (A113G), and A113 and V291 are both replaced by glycine and leucine, respectively (A113G/V291L), have been constructed. With the expectation of the opposite effect, we also prepared the mutant PheDH, in which G114 was changed to alanine (G114A-PheDH) by PCR mutagenesis. Furthermore, to alter the specificities of LeuDH toward GluDH or LysDH, L40, corresponding to K89 in GluDH, whose ε-amino group interacts with the γ-carboxyl group of the substrate L-glutamate by hydrogen bonding, has been mutated to lysine (L40K) or aspartate (L40D). V294 of LeuDH, which is equivalent to S379 in GluDH that interacts with the substrate glutamate by hydrogen bonding, has been replaced as well by serine for making double mutants (L40K/V294S, L40D/V294S). In these variants, the hydrophilicity of the substrate side-chain binding pocket increases, and the interaction of the substrate with the polar side chain is expected. All seven mutant enzymes were expressed and purified from the crude extract of recombinant E. coli cells as the wild-type enzymes. After the purification, all mutant enzymes 9 exhibited a single prominent band on the SDS-PAGE gel with molecular sizes equivalent to the wild-type LeuDH or PheDH (data not shown). 3.2. CD spectra of mutants CD spectra of the wild-type and six mutant enzymes of LeuDH were measured to see whether the global conformations were changed by the mutation. All the mutant enzymes except A113G/V291L showed CD spectra practically identical with that of the wild-type enzyme in the 200-250 nm region (Fig. 2), indicating that the wild-type and these mutant enzymes contain very similar secondary structures. The decreased CD band of A113G/V291L in this region indicates the decreased secondary structure content and some slight conformational changes in the overall structure of this mutant. On the other hand, the spectra of mutant enzymes in the 260-290 nm region, which reflects the environment of aromatic residues, were significantly different from that of the wild-type enzyme (Fig. 2A). The CD band of the wild-type LeuDH in this region is mainly due to the sole tryptophan (W46), which is predicted to be located in the vicinity of the active site [37]. The decreases in the near-UV CD bands of A113G and A113G/V291L indicate that the environment around W46 of these mutants became more hydrophobic than that of the wild-type enzyme, as in the case of mutations of the conserved glycine residues adjacent to the catalytic K80 [37]. On the contrary, the spectrum of the L40D/V294S mutant in this region, which is more negative than that of the wild-type, is indicative of a more hydrophilic environment around W46 than in the wild-type enzyme. CD spectrum of the G114A mutant of PheDH appeared essentially identical to that of the wild-type enzyme (data not shown). Thus, the G114A mutant appears to have an identical conformation to that of the wild-type PheDH. 10 3.3. Catalytic properties of A113G and A113G/V291L mutants of LeuDH and G114A mutant of PheDH Both single and double mutants of LeuDH prepared to mimic PheDH showed a marked decrease in activities towards the aliphatic substrates compared to the wild-type enzyme. The kcat values of the single mutant A113G and the double mutant A113G/V291L decreased by about 20 and 60 folds for L-leucine, and by about 3 and 10 folds for α-keto-isocaproate, respectively (Table 2). The substrate specificities of A113G and A113G/V291L were examined in the oxidative deamination with various amino acids and in the reductive amination with various α-keto acids as substrate at 10 mM. The results are compared with those of LeuDH and PheDH in Table 1. Both of the mutant enzymes have broader substrate specificities than the wild-type LeuDH. The mutants act on, in addition to the preferred substrate of LeuDH (short branched-chain amino acids including L-leucine, L-isoleucine, Lvaline, and their keto-analogs), poor substrates of LeuDH such as L-norleucine, L-norvaline, L-methionine, L-ethionine, and α-keto-γ-methylthiobutyrate (long-chain aliphatic substrates). Furthermore, these two mutants also utilize L-phenylalanine and phenylpyruvate with over 10% relative activity for the preferred substrate L-leucine and its keto-analog, while the wildtype enzyme does not respond at all to the substrates with aromatic side chains. This could be attributed to the fact that the methyl side chain of A113 in LeuDH, which sterically interferes with binding of the substrate aromatic ring, was removed by substitution to glycine. Table 2 summarizes the kcat and Km values of the mutant and wild-type enzymes of LeuDH and PheDH. Although the kcat value for L-leucine and its keto analog of the LeuDH single mutant decreased considerably, the values for other aliphatic substrates did not change so much. In addition, the Km values of A113G for aliphatic substrates are similar to the 11 corresponding values of the wild-type LeuDH. This result suggests that there is no significant conformational change in the active site of the mutant upon replacement of A113 with glycine. As expected from the molecular modeling, the single mutant showed PheDH activities, even though they were merely 4% and 11%, respectively, of those of the wild-type PheDH in the oxidative deamination of L-phenylalanine and reductive amination of phenylpyruvate. The Km values of the A113G mutant for L-phenylalanine and its keto analog were also over 100 times larger than the values of the wild-type PheDH. The low affinities of A113G for both Lphenylalanine and phenylpyruvate lead to a conclusion that the side-chain methyl group of A113 in LeuDH is critical for discrimination between substrates with or without an aromatic side chain but its absence does not contribute for increasing the affinity for aromatic substrates. In contrast with the single mutant, the kcat values of the A113G/V291L double mutant of LeuDH greatly decreased with all the aliphatic substrates examined. Furthermore, the Km values for all amino acid and α-keto acid substrates increased over 10 folds, although there was not such a large change in the Km values for the coenzymes (NAD+ and NADH). These results are consistent with the significantly altered CD properties of the double mutant in the UV region, as described above. It is plausible that the double mutation has caused some perturbation of local conformations in the substrate side-chain binding region. The G114A mutant of PheDH showed broader substrate specificity than the wildtype enzyme, and the relative activity for the aliphatic substrates (L-isoleucine, L-norvaline, L-methionine, α-keto-β-methylvalerate, and α-ketobutyrate) increased markedly (Table 1). Despite the fact that the A113G mutant of LeuDH is considerably active with phenylalanine and phenylpyruvate likely due to the removal of the side-chain methyl group, the G114A mutant of PheDH still retains a low activity toward aromatic substrates (4.2% of kcat of the 12 wild-type PheDH for L-phenylalanine, Table 2). The broad substrate specificity of this mutant is consistent with the specificities of the corresponding G124A and G124A/L307V mutants of B. sphaericus PheDH, which also retained about 5% of activities for substrates with an aromatic side chain [32]. Comparing the Km values of the G114A mutant with those of the wild-type PheDH, we note that the values for aliphatic substrates are unchanged, although those for both L-phenylalanine and phenylpyruvate increased by 160 times. Therefore, we conclude that the affinity for aromatic substrates decreases specifically by introducing a mutation in this position. The greatly decreased activity and the affinity towards aromatic substrates with G114A indicate that G114 in PheDH is indeed critical for recognition of aromatic substrates. 3.4. Catalytic properties of L40K and L40K/V294S mutants of LeuDH The substrate specificities of the mutants that were designed to mimic the active site of GluDH were also studied in both the oxidative deamination and reductive amination reactions (Table 3). The specificity of the single mutant L40K, in which L40 of LeuDH has been replaced by lysine, corresponding to K89 in GluDH that plays a role in anchoring the γcarboxyl group of the substrate L-glutamate, did not change greatly from that of the wild-type LeuDH. It could react neither with L-glutamate nor with α-ketoglutarate as substrates. In contrast, the double mutant L40K/V294S with an additional mutation of V294 to serine (equivalent to S379 in GluDH, interacting with the substrate side chain) has a weak GluDH activity, as expected from the molecular modeling. The double mutant of LeuDH acts on polar amino acids such as L-asparagine, L-aspartate, L-glutamine, and L-glutamate, in addition to the preferred substrates of the wild-type enzyme. 13 The kinetic parameters of the single and double mutants of LeuDH in both deamination and amination reactions are presented in Table 4. The single mutant showed reduced kcat values as compared to those of the double mutant having similar values with the wild-type enzyme. The Km values of the single mutant for L-isoleucine, L-norleucine, α-ketoiso-caproate, and α-keto-β-methylvalerate increased by more than 10 folds as compared to the wild-type. The kinetic parameters of L40K are similar to those obtained with the K89L mutant of GluDH, prepared in a manner opposite to the present studies [31]. Both of the single mutants (L40K of LeuDH and K89L of GluDH) displayed large reductions in the original catalytic activities and failed to reverse the substrate specificities. In the L40K mutant, the positively charged lysine side chain has been introduced, but the hydrophobic valine remains in position 294. The K89L mutant of GluDH has oppositely the L89/S379 pair. Presumably, these incompatible pairings of side chains cannot be stabilized by binding of the normal or alternative substrates. In marked contrast to the single mutant, the double mutant of LeuDH (L40K/V294S) showed kcat values even similar to those of the wild-type enzyme (except for L-leucine). However, the double mutant had high Km values for the preferred substrates of LeuDH, like the single mutant. Although the single and double mutants had slight activities toward Lglutamate and α-ketoglutarate, the catalytic rates with the two substrates were unsaturable with their concentrations (data not shown); thus the Km values for L-glutamate and αketoglutarate could not be obtained, and the kcat values, determined with the substrate at 50 mM, shown in Table 4 are a lower limit. The high Km values of the double mutant for aliphatic substrates are indicative of the increased hydrophilicity of the pocket that is advantageous for binding an acidic substrate, but further fine tuning of the substrate-binding pocket is certainly required to realize the high affinity for L-glutamate. 14 3.5. Catalytic Properties of L40D and L40D/V294S mutants of LeuDH We constructed another set of mutant enzymes of LeuDH in order to confer the activity for the substrates with a positively charged side chain (L-lysine and L-arginine). As shown in Table 3, the substrate specificities of the single (L40D) and double (L40D/V294S) mutants for aliphatic amino acids were virtually unaffected as compared with the wild-type enzyme. These mutants, however, displayed entirely new activities for L-lysine and Larginine, even in the single mutant (L40D), unlike in the case of the L40K mutant described above. However, the substitution of L40 with aspartate caused marked reductions of kcat values for aliphatic substrates (lower by ~100 folds) compared with those of the wild-type LeuDH (Table 4). Moreover, the Km values for these preferred substrates increased much; for example, the values for L-leucine, L-isoleucine, and L-norleucine were too high to determine. The large reduction of activities of the mutants for aliphatic substrates seems to be mainly due to the increased Km values. The kcat values of the single and double mutants for deamination of L-lysine were 0.029 and 0.45 s–1, which were only 0.06 and 0.9%, respectively, of that for L-leucine of the wild-type LeuDH. Nevertheless, the Km values of the mutants for L-lysine were only one order of magnitude higher than those for aliphatic substrates of the wild-type enzyme. These kinetic properties, together with the CD spectrum of the L40D/V294S mutant, suggest that the hydrophilicity of the substrate side-chain binding pocket, enhanced by the substitutions of L40 by aspartate and V294 by serine, is important in permitting the binding of a substrate with a positively charged side chain such as L-lysine. 15 4. Conclusions Enzymes belonging to the amino acid dehydrogenase superfamily have already been used to produce various amino acids in commercially feasible quantities. Understanding the molecular mechanisms for the distinct substrate specificities displayed by these enzymes would enable us to engineer the substrate specificities by site-directed mutagenesis and further facilitate their applications to the enzymatic production or detection of novel amino acids. In the studies herein reported, we constructed three types of mutant enzymes of LeuDH to mimic the substrate specificities of PheDH, GluDH, and LysDH by protein engineering. The kinetic parameters obtained with these mutants clearly showed that A113, V294, L40, and V294 in LeuDH are the key residues involved in recognition of the substrate side chain. A113 in LeuDH and the corresponding G114 in PheDH control the volume of the side-chain binding pocket and play a critical role in discrimination of the bulkiness of the side chain. Emergence of GluDH and LysDH activities in the L40K/V294S and L40D/V294S mutants, respectively, indicates that the two residues in the active sites of the amino acid dehydrogenases are important for discrimination of the hydrophobicity/polarity/charge of aliphatic substrate side chains. Acknowledgements This work was supported by a Grant-in-Aid for Encouragement of Young Scientists (to K.K., No. 07780537) from the Ministry of Education, Science, Sports and Culture of Japan. 16 References [1] T. Ohshima, H. Misono and K. Soda, J. Biol. Chem., 253 (1978) 5719 [2] M.W. Zink and B.D. Sanwal, Arch. Biochem. Biophys., 99 (1962) 72 [3] J. Hermier, J.M. Lebeault and C. Zevako, Bull. Soc. Chim. Biol., 52 (1970) 1089 [4] J. Hermier, M. Rosseau and C. Zevako, Ann. Inst. Pasteur Paris, 118 (1970) 611 [5] R.L. Hanson, J. Singh, T.P. Kissick, R.N. Patel, L.J. Szarka and R.H. Mueller, Bioorg. Chem., 18 (1990) 371 [6] A.S. Bommarius, M. Schwarn and K. Drauz, J. Mol. Catal. B: Enzymatic, 5 (1998) 1 [7] K.C. Dooley, Clin. Biochem., 25 (1992) 271 [8] G. Livesey and P. Lund, Methods Enzymol., 166 (1988) 3 [9] A. Yamaguchi, Y. Mizushima, M. Fukushi, Y. Shimizu, Y. Kikuchi and N. Takasugi, Screening, 1 (1992) 49 [10] S.H. Mudd, H.L. Levy and F. Skovby, in C.R. Sciver, A.L. Beudet, W.S.Sly and D. Valle (Ed.), The Metabolic Basis of Inherited Disease, McGraw-Hill, New York, 1989, p. 693 [11] S. Nagata, K. Tanizawa, N. Esaki, Y. Sakamoto, T. Ohshima, H. Tanaka and K. Soda, Biochemistry, 27 (1989) 9056 [12] J.K. Teller, R.J. Smith, M.J. McPherson, P.C. Engel and J.R. Guest, Eur. J. Biochem., 206 (1992) 151 [13] M.J. McPherson and J.C. Wootton, Nucleic Acid Res., 11 (1983) 5257 [14] H. Takada, T. Yoshimura, T. Ohshima, N. Esaki and K. Soda, J. Biochem., 109 (1991) 371 [15] N. Okazaki, Y. Hibino, Y. Asano, M. Ohmori, N. Numao and K. Kondo, Gene, 63 (1988) 337 [16] A. Leiser, A. Birch and J.A. Robinson, Gene (Amst.), 177 (1996) 217 17 [17] L. Tang and C.R. Hutchinson, J. Bacteriol., 175 (1993) 4176 [18] K.L. Britton, P.J. Baker, P.C. Engel, D.W. Rice and T.J. Stillman, J. Mol. Biol., 234 (1993) 938 [19] S.E.H. Syed, P.C. Engel and D.M. Parker, Biochim. Biophys. Acta, 1115 (1991) 123 [20] T. Ohshima, N. Nishida, S. Bakthavatsalam, K. Kataoka, H. Takada, T. Yoshimura, N. Esaki and K. Soda, Eur. J. Biochem., 222 (1994) 305 [21] N.D. Priestley and J.A. Robinson, Biochem. J., 261 (1989) 853 [22] Y. Asano, A. Nakazawa and K. Endo, J. Biol. Chem., 262 (1987) 10346 [23] T. Ohshima, H. Takada, T. Yoshimura, N. Esaki and K. Soda, J. Bacteriol., 173 (1991) 3943 [24] P.J. Baker, K.L. Britton, P.C. Engel, G.W. Farrants, K.S. Lilley, D.W. Rice and T.J. Stillman, Proteins, 12 (1992) 75 [25] P.J. Baker, A.P. Turnbull, S.E. Sedelnikova, T.J. Stillman and D.W. Rice, Structure, 3 (1995) 693 [26] J.L. Vanhooke, J.B. Thoden, N.M.W. Brunhuber, J.S. Blanchard and H.M. Holden, Biochemistry, 38 (1999) 2326 [27] K. Kataoka, H. Takada, K. Tanizawa, T. Yoshimura, N. Esaki, T. Ohshima and K. Soda, J. Biochem., 116 (1994) 931 [28] T. Oikawa, K. Kataoka, Y. Jin, S. Suzuki and K. Soda, Biochem. Biophys. Res. Commun., 280 (2001) 1177 [29] T.J. Stillman, P.J. Baker, K.L. Britton and D.W. Rice, J. Mol. Biol., 234 (1993) 1131 [30] P.J. Baker, M.L. Waugh, X.-G. Wang, T.J. Stillman, A.P. Turnbull, P.C. Engel and D.W. Rice, Biochemistry, 36 (1997) 16109 18 [31] X.-G. Wang, K.L. Britton, P.J. Baker, S. Martin, D.W. Rice and P.C. Engel, Protein Eng., 8 (1995) 147 [32] Y.K. Seah, K.L. Britton, P.J. Baker, D.W. Rice, Y. Asano and P.C. Engel, FEBS Lett., 370 (1995) 93 [33] Y.K. Seah, K.L. Britton, D.W. Rice, Y. Asano and P.C. Engel, Biochemistry, 41 (2002) 11390 [34] T.A. Kunkel, J.D. Roberts and R.A. Zakour, Methods Enzymol., 154 (1987) 367 [35] K. Kataoka, K. Tanizawa, T. Fukui, H. Ueno, T. Yoshimura, N. Esaki and K. Soda, J. Biochem., 116 (1994) 1370 [36] K. Kataoka, M. Nakai, K. Yamaguchi and S. Suzuki, Biochem. Biophys. Res. Commun., 250 (1998) 409 [37] T. Sekimoto, T. Fukui and K. Tanizawa, J. Biochem., 116 (1994) 176 19 Figure legends Fig. 1. Sequence alignment of amino acid dehydrogenases using structural information from B. sphaericus LeuDH and C. symbiosum GluDH. Residues L40, A113, V291, and V294 of B. stearothermophilus LeuDH and their corresponding residues are shown with white letters in black background. Fig. 2. CD spectra of the wild-type and mutant enzymes of LeuDH. The CD spectra were recorded at 25 ºC in 10 mM potassium phosphate buffer (pH 7.2). (A) Spectra of the wildtype (solid line), A113G (broken line), and A113G/V291L (dotted line) mutant enzymes. (B) Spectra of the wild-type (solid line), L40K (broken line), and L40K/V294S (dotted line) mutant enzymes. (C) Spectra of the wild-type (solid line), L40D (broken line), and L40D/V294S (dotted line) mutant enzymes. The unit of [θ] is mdeg•cm2•dmol–1. 20 Table 1. Substrate specificities of wild-type and mutant enzymes a Substrate (10 mM) Wild type Relative activity (%) LeuDH PheDH A113G AG/V291L Wild type G114A Deamination L-Leucine L-Isoleucine L-Valine L-Norleucine L-Norvaline L-Methionine L-Ethionine L-Alanine L-Phenylalanine L-Tyrosine L-Tryptophan L-Histidine L-Glutamate 100 54 39 14 56 0.7 0 0 0 0 0 0 0 100 550 62 530 110 81 35 0 15 0 0 0.4 0 100 150 13 110 23 19 12 0 11 0 0 0 0 3.9 0.4 1.3 6.3 2.1 2.2 4.0 0.4 100 40 1.2 0.2 0 42 90 39 66 72 72 11 3.5 100 5.4 0 1.3 0 Amination α-Keto-iso-caproate α-Keto-β-methylvalerate α-Ketovalerate α-Keto-γ-methylthiobutyrate α-Ketobutyrate Phenylpyruvate p-Hydroxyphenylpyruvate 100 100 86 15 47 0 0 100 180 102 130 5.1 17 0 100 67 19 36 0.6 11 0 47 16 37 55 5.5 100 80 151 174 224 102 104 100 0 a Specific activities of the wild-type and mutant enzymes of LeuDH and PheDH for various substrates are shown as % activities relative to the values for L-leucine and L-phenylalanine, respectively, taken as 100. 21 a Table 2. Steady-state kinetic parameters of wild-type and mutant enzymes LeuDH A113G A113G/V291L PheDH G114A –1 kcat (s ) L-Leucine L-Isoleucine L-Norleucine L-Norvaline L-Phenylalanine α-Keto-iso-caproate α-Keto-β-methyl valerate α-Ketocaproate Phenylpyruvate Km (mM) L-Leucine L-Isoleucine L-Norleucine L-Norvaline L-Phenylalanine α-Keto-iso-caproate α-Keto-β-methyl valerate α-Ketocaproate Phenylpyruvate NAD+ (µM) NADH (µM) a 50 ± 3.0 28 ± 0.77 1.3 ± 0.03 13 ± 0.15 0 2.8 ± 0.08 0.74 ± 0.010 0.20 ± 0.006 0.12 ± 0.002 23 ± 0.53 1.4 ± 0.037 0.24 ± 0.02 0.22 ± 0.006 18 ± 0.04 0.60 ± 0.018 1.9 ± 0.005 0.16 ± 0.004 7.7 ± 0.19 0.29 ± 0.008 0.59 ± 0.006 0.21 ± 0.004 0.80 ± 0.004 1.4 ± 0.049 23 ± 0.27 0.97 ± 0.012 280 ± 22 94 ± 1.9 29 ± 0.35 36 ± 0.49 37 ± 0.48 280 ± 4.1 235 ± 5.2 37 ± 0.52 26 ± 0.22 46 ± 0.24 110 ± 2.4 140 ± 2.8 24 ± 0.91 79 ± 1.1 61 ± 0.49 9.9 ± 1.2 200 ± 1.3 36 ± 0.46 0 22 ± 0.70 5.1 ± 0.50 1.4 ± 0.10 35 ± 2.5 0.12 ±0.006 0.082 ± 0.002 2.4 ± 0.15 7.8 ± 0.49 33 ± 1.8 0.081 ± 0.007 0.082 ± 0.003 4.4 ± 0.21 4.1 ± 0.026 24 ± 1.6 0.19 ± 0.002 0.17 ± 0.008 27 ± 1.8 70 ± 8.6 0.50 ± 0.016 0.18 ± 0.006 31 ± 1.2 66 ± 7.1 0.10 ± 0.003 4.4 ± 0.19 7.8 ± 0.29 ─ 0.88 ± 0.15 1.7 ± 0.11 30 ± 2.7 3.8 ± 0.16 9.4 ± 0.59 69 ± 5.1 4.7 ± 0.34 2.0 ± 0.10 28 ± 2.9 ─ 7.1 ± 1.1 9.9 ± 1.2 12 ± 0.45 2.5 ± 0.11 0.065 ± 0.003 16 ± 0.42 4.1 ± 0.16 5.2 ± 0.067 4.4 ± 0.11 11 ± 0.47 63 ± 7 76 ± 1 280 ± 13 170 ± 10 50 ± 20 35 ± 6 42 ± 2 17 ± 7 83 ± 1 27 ± 7 Steady-state kinetic parameters were determined by varying the concentration of the substrate to be measured in the presence of fixed concentrations of the cofactor and cosubstrate (1.25 mM NAD+ in the oxidative deamination and 0.1 mM NADH and 1 M ammonia in the reductive amination). 22 Table 3. Substrate specificities of wild-type and mutant enzymes of LeuDH a Substrate (10 mM) Wild type Relative activity (%) L40K L40K/V294S L40D L40D/V291S Deamination L-Leucine L-Isoleucine L-Valine L-Norleucine L-Norvaline L-Methionine L-Ethionine L-Alanine L-Phenylalanine L-Histidine L-Asparagine L-Glutamine L-Aspartate L-Glutamate L-Lysine L-Arginine 100 54 39 14 56 0.7 0 0 0 0 0 0 0 0 0 0 100 273 81 28 29 2.6 0.5 0.04 0 0.02 0 0 0 0 0 0 100 187 26 17 24 3.6 0.2 0 0 0 0.1 0.4 0.04 0.1 0 0 100 212 64 35 37 3.0 0 0 0 0 0 0 0 0 9.2 1.8 100 358 45 27 33 4.2 0 0 0 0 0 0 0 0 38 1.1 Amination α-Keto-iso-caproate α-Keto-β-methylvalerate α-Ketovalerate α-Ketobutyrate 100 100 86 47 100 86 100 58 30 2.7 100 92 52 6.1 100 63 a b ND b ND Specific activities of the wild-type and mutant enzymes of LeuDH for various substrates are shown as % activities relative to the values for L-leucine, taken as 100. b b ND b ND Not determined. 23 a Table 4. Steady-state kinetic parameters of wild-type and mutant enzymes of LeuDH Wild type L40K L40K/V294S L40D L40D/V291S –1 kcat (s ) L-Leucine L-Isoleucine L-Norleucine L-Glutamate L-Lysine 50 ± 3.0 28 ± 0.77 1.3 ± 0.03 0 0 α-Keto-iso-caproate 280 ± 22 α-Keto-β-methyl 280 ± 4.1 valerate α-Ketoglutarate 0 Km (mM) L-Leucine L-Isoleucine L-Norleucine L-Glutamate L-Lysine α-Keto-iso-caproate α-Keto-β-methyl valerate α-Ketoglutarate a 5.1 ± 0.50 2.4 ± 0.15 4.4 ± 0.21 3.8 ± 0.038 1.8 ± 0.41 1.9 ± 0.04 0 0 9.8 ± 0.054 17 ± 0.46 1.7 ± 0.03 b 0.02 0 b b 0.21 b 0.49 b 0.070 0 0.029 ± 0.001 0.80 b 2.8 b 0.22 0 0.45 ± 0.03 95 ± 1.7 270 ± 5.9 9.9 ± 0.051 47 ± 0.50 82 ± 1.6 280 ± 3.5 1.9 ± 0.024 30 ± 0.47 b 0 0 0.57 5.2 ± 3.7 64 ± 1.8 110 ± 3.0 ─ ─ 26 ± 0.52 24 ± 1.5 27 ± 1.1 c ND ─ ─ ─ 0 c c ND c ND c ND ND c ND c ND ─ ─ 64 ± 3.9 72 ± 5.0 0.88 ± 0.15 46 ± 3.2 42 ± 3.3 260 ± 22 120 ± 11 3.8 ± 0.16 49 ± 3.8 84 ± 6.7 47 ± 2.4 110 ± 14 ─ c ND ─ ─ ─ Steady-state kinetic parameters were determined by varying the concentration of the substrate to be measured in the presence of fixed concentrations of the cofactor and cosubstrate (1.25 mM NAD+ in the oxidative deamination and 0.1 mM NADH and 1 M ammonia in the reductive amination). b c Determined for the substrate at 50 mM. Not determined due to the catalytic rates,being unsaturable with the substrate concentrations. 24 !c LeuDH B. sphaericus LeuDH B. stearo. LeuDH T. intermedius PheDH T. intermedius PheDH B. sphaericus GluDH E. coli GluDH C. symbiosum GPALGGARMF GPALGGTRMW 37 GPALGGMRMW 38 GPALGGCRMI 48 GPALGGTRMY 90 GPYKGGMRFH 86 GPYKGGLRFA 37 37 !e "4 RYITAEDVGTTVSDM RYITAEDVGTTVADM 109 RYITAEDVGTTVEDM 110 RFYTGTDMGTNPEDF 120 RFYTGTDMGTTMDDF 163 TDVPAGDIGVGGREV 160 IDVPAGDLGVGAREI 109 109 !c !e Fig.1 25 "8 "12 YVINAGGVINVADELYGY YVINAGGVINVADELYGY 284 YVINAGGVINVADELLGY 287 YLVNAGGLIQVADELEGP 270 YLVNAGGLIQVADELEGP 371 KAANAGGVATSGLEMAQN 370 KAVNAGGVLVSGFEMSQN 284 284 "15a "15b 1 A 0 0 -2 -1 -4 -2 -6 -3 B 0 0 -2 -1 -4 -2 -6 -3 C 0 0 -2 -1 -4 -2 -6 200 220 240 260 280 300 Wavelength (nm) 26 -3 320 [!]x10-2 [!]x10-4 2