To determine which amino acid residues are essential for the catalytic activity of soybean P-amylase, deoxyoligonucleotide site-directed mutagenesis was employed against aspartyl, glutamyl, and cysteinyl residues located in highly conserved regions found in P-amylase family to date. Both substitution of aspartic acid at position 101 and that of glutamic acid at position 186 of the enzyme by neutral and acidic amino acids, respectively, led to the complete elimination of activity, but did not induce any significant changes in circular dichroic spectra or the binding affinity for cyclomaltohexaose, a substrate analogue.Taking account of the results obtained here, the above two amino acid residues are involved in the catalytic site of soybean P-amylase. The replacement of glutamic acid at position 345 decreased activity to below 6% of the non-mutant level, implying that this residue may also play a crucial role in /I-amylase activity, although it may not be involved at the catalytic site itself. In contrast, substitution of cysteinyl residue at position 95 by a serinyl residue led to a drastic reducing of the optimal temperature (from 50°C to 30"C), suggesting that this cysteinyl residue is responsible for the thermal stability of the enzyme.8-Amylase (1 ,4-a-D-glucan maltohydrolase) hydrolyzes a-l,4-glucosidic linkage of starch and glycogen with liberation of P-anomeric maltose from the non-reducing ends. This enzyme is found in many plant and bacteria species. Although the physicochemical properties of the enzymes from sweet potato, barley, and soybean have been well investigated [l], the catalytic mechanism of the enzyme has not yet been elucidated. Comparison of deduced sequences of P-amylases originating from plant and bacteria revealed three highly conserved sequences which very likely are involved in the segment assembly of the catalytic site of /I-amylase [2-41. In the case of a-amylase (1,4-a-~-glucan glucanohydrolase) that catalyzes the hydrolysis of 1,4-a-D-glucans, the aspartyl and glutamyl residues contained in highly conserved regions have been identified as catalytic residues by X-ray crystallography [5 -71 and by site-directed mutagenesis [8, 91. Although the action pattern of P-amylase differs from that of a-amylase and there is no significant similarity between the primary sequences, all of the aspartyl and glutamyl residues located in the conserved regions of P-amylase family can be considered possible catalytic residues. Recently, the results of affinity labeling of soybean P-amylase with 2,3-epoxy propyl a-D-glucopyranoside suggested that Glu186 was involved in the catalytic site [lo]. In addition, since the enzyme was inactivated by sulfhydryl reagents such as iodoacetamide, N-ethylmaleimide and mercuribenzoate [ l , 111, the cysteinyl residues located in the regions may also participate in the catalytic reaction.

The cDNA clones corresponding to soybean P-amylase mRNA were isolated and sequenced. The cDNA contained an open-reading frame composed of 496 amino acids. The comparison of the amino acid sequence deduced from the cDNA with the N-terminal peptide sequence from mature enzyme proved that P-amylase had no leader sequence. Employing the cDNA, the P-amylase was directly synthesized in Escherichia coli by the expression vector pKK233-2 controlled by the tac promoter. The enzyme activity detected in E. coli lysate drastically increased with a lower cultivation temperature, and the total activity and specific activity of the enzyme in E. coli lysate cultured at 13°C was 130-fold and 280-fold, respectively, the value at 37°C. The enzyme produced in E. coli was purified by the affhity column chromatography of cyclomaltohexaose-immobilized Sepharose 6B. Employing the established expression and purification system of the enzyme, the functional ionizable groups in the active site were searched. His93, involving an imidazole, and Asp348, involving a carboxylate, in the highly conserved regions within the P-amylases were replaced by Arg (H93R) and Asn (D348N) by site-directed mutagenesis, respectively. All P-amylases, including the non-mutant and mutant @-amylases, produced in E. coli exhibited lower V,, values than that of P-amylase isolated conventionally from soybean seeds. Especially the V,,, value of [H93R]P-amylase was reduced drastically compared to that of the non-mutant; however, none of them lost their enzyme activities completely. Therefore, neither His93 nor Asp348 may participate in the catalytic reaction directly. 7] and Clostridiurn themzosulfurogenes [ 81 were cloned and sequenced. The highly conserved amino acid sequences found among them suggest that those regions may assemble the active center of P-amylase reaction [9, 101. Although the crystallographic study of soybean P-amylase is in progress [ll], the relationship between the primary structure and the catalytic reaction still remains unclear. The thiol group of the cysteine residue [12, 131 or two functional ionizable groups with pKa 3.5, such as a carboxylate of a Glu or Asp residue, and pK, 8.2, such as an &-amino group of Lys or an imidazole group of His [14], were speculated to be involved in the active site. Recently, using affinity labeling of soybean Pamylase with 2,3-epoxy propyl a-D-glucopyranoside, Glul86 was shown to be involved in the active site [15]. Also the direct contributions of cysteine residues to the catalytic reacCorrespondence to C . Fukazawa, Genetic Engineering Laboratory, National Food Research Institute, Ministry of Agriculture, Forestry and Fisheries, Kannondai, Tsukuba, Ibaraki 305, JapanFar: +81 0298 38 7996.Enzymes. a-Amylase, 1,4-a-r>-glucan glucanohydrolase (EC 3.2.1.1); P-amylase, 1,4-a-D-ghCan maltohydrolase (EC 3.2.1.2).Note. The novel nucleotide sequence data published here have been deposited with the GenBank EMBL sequence data bank and are available under the accession number X71419. The novel amino acid sequence d...

No abstract