Christensen U, Olsen K, Stoffer B B, Svensson B
Chemical Laboratory IV, University of Copenhagen, Denmark.
Biochemistry. 1996 Nov 26;35(47):15009-18. doi: 10.1021/bi9608323.
Glucoamylase (1,4-alpha-glucan glucohydrolase, EC 3.2.1.3) from Aspergillus, of which the 3D structure is known, releases beta-D-glucose from the non-reducing ends of starch and other related oligo and polysaccharides, cleaving the alpha-1,4-bond positioned between subsites 1 and 2 in the enzyme-substrate complex. The presteady and steady state kinetics of two of the existing mutants, Glu180-->Gln and Asp176-->Asn, are presented here. The kinetic results are analyzed according to two reaction models: One suggested previously [Olsen, K., Svensson, B., & Christensen, U. (1992) Eur. J. Biochem. 209, 777-784], which contains three consecutive steps of the reaction, and one generally accepted and used in calculations of subsite energies [Hiromi, K. (1970) Biochem. Biophys. Res. Commun. 40, 1-6], which assumes important non-productive binding and identical values of the intrinsic catalytic constant independent of the chain length of the substrate. It is found that glucoamylase shows kinetics in accordance with a consecutive three-step mechanism, in which the formation of the Michaelis complex occurs in two steps and is followed by a slow catalytic step and fast dissociation of the products with no accumulation of enzyme-product complexes. The kinetics, however, are not in accordance with the model generally used in subsite energy calculations. Thus the kinetic model on which very low values of subsite 1 and high values of subsite 2 interaction energies have been based is not correct. A greater importance of subsite 1 interactions than has hitherto been anticipated is indicated. The results of the Glu180-->Gln mutant show weak overall binding, which stems from large effects on the formation of the Michaelis complex in the second step of the reaction, but no or rather small effects on the initial association of enzyme and substrate, except for maltose. The mutant further shows effective catalysis. A hydrogen bond of the side chain carboxylate of Glu180 with the 2-OH of the sugar ring at subsite 2 is an expected important interaction of the Michaelis complex, as seen from the 3D structures of stabile enzyme-inhibitor complexes. Apparently this bond is established in the second reaction step. It is indicated that subsite 1 and 3 interactions to a great extent govern the initial association. In accordance with a dynamic role of Glu180, structural energy minimization calculations show a flexibility of the gamma-carboxylate of Glu180. The side chain of Asp176 participates in a hydrogen-bonding network also involving the backbone of Glu180 and Glu179, the catalytic acid. Compared with the wild-type enzyme, the Asp176-->Asn mutant shows no significant changes in binding. The catalytic rate is, however, markedly reduced. Apparently changes in the hydrogen bonding network of Asp176 are of importance in the rate-determining catalytic step, but not in the substrate binding steps. Structural energy minimization calculations on the Asp176-->Asn mutant, however, do not confirm this assumption.
来自曲霉的葡糖淀粉酶（1,4-α-葡聚糖葡糖水解酶，EC 3.2.1.3），其三维结构已知，可从淀粉及其他相关寡糖和多糖的非还原端释放β-D-葡萄糖，切断酶-底物复合物中亚位点1和2之间的α-1,4-键。本文给出了现有两个突变体Glu180→Gln和Asp176→Asn的预稳态和稳态动力学。根据两种反应模型对动力学结果进行分析：一种是先前提出的[奥尔森，K.，斯文森，B.，&克里斯蒂安森，U.（1992）欧洲生物化学杂志209，777 - 784]，其中包含反应的三个连续步骤；另一种是在亚位点能量计算中普遍接受和使用的[广见，K.（1970）生物化学与生物物理研究通讯40，1 - 6]，该模型假定存在重要的非生产性结合且内在催化常数的值与底物链长无关。研究发现葡糖淀粉酶的动力学符合连续三步机制，其中米氏复合物的形成分两步进行，随后是一个缓慢的催化步骤以及产物的快速解离，且没有酶-产物复合物的积累。然而，其动力学并不符合亚位点能量计算中通常使用的模型。因此，基于亚位点1值极低和亚位点2相互作用能极高所建立的动力学模型是不正确的。这表明亚位点1相互作用的重要性比迄今预期的更大。Glu180→Gln突变体的结果显示总体结合较弱，这源于对反应第二步中米氏复合物形成的较大影响，但对酶与底物的初始结合影响不大或相当小，除了麦芽糖。该突变体还表现出有效的催化作用。从稳定的酶-抑制剂复合物的三维结构可以看出，Glu180侧链羧酸盐与亚位点2糖环的2-OH形成的氢键是米氏复合物预期的重要相互作用。显然，这个键是在第二个反应步骤中形成的。这表明亚位点1和3的相互作用在很大程度上决定了初始结合。与Glu180的动态作用一致，结构能量最小化计算显示Glu180的γ-羧酸盐具有灵活性。Asp176的侧链参与了一个氢键网络，该网络还涉及Glu180和催化酸Glu179的主链。与野生型酶相比，Asp176→Asn突变体在结合方面没有显著变化。然而，催化速率明显降低。显然，Asp176氢键网络的变化在决定速率的催化步骤中很重要，但在底物结合步骤中并非如此。然而，对Asp176→Asn突变体的结构能量最小化计算并未证实这一假设。
