Models for Galactose Oxidase (GOase) Based on Combinatorial Peptoid Libraries

金属酶的低分子量模型通常是根据活性中心的官能团、金属离子、辅酶等的空间排列信息设计的(理想情况下，这些信息是通过X射线晶体分析获得的)。这种方法在很大程度上忽略了蛋白质基质的影响，特别是它们的动力学、局部偶极矩、对金属络合物聚集的抑制等。因此，许多仿生的低分子过渡金属络合物能够再现相应酶的结构和光谱特性，但大多数都没有表现出显著的催化活性。在我们的方法中，我们还从给定的酶结构中推断出哪些配体结合到催化活性的金属中心。然而，在下一步，合成的组合配位库将金属配位基团放置在寡肽的不同位置。通过适当的分析，确定了文库的催化活性成员。换句话说，该系统在构成和动态方面都被赋予了相当大的自由度，只有那些具有催化活性的实体才被进一步研究。目前的项目旨在确定半乳糖氧化酶的功能模型，即能够将有氧的醇氧化为相应的羰基化合物的仿生铜配合物。配体库将在固相上制备，编码将通过irori射频标记技术进行。文库的设计是基于树枝指甲真菌酶的X射线晶体结构：组氨酸、酪氨酸和模拟野生型酶的Cys228-Tyr272结合物的非天然氨基酸将被用作金属结合实体。非天然的Cys-Tyr模拟物将从适当的手性分子中以对映体纯形式制备。GOASE活性的分析将基于现有的检测过氧化氢的酶方法，或基于产物羰基化合物的染料形成反应。这些可用于高通量筛选的反应性分析，将辅之以“经典”(手性)气相色谱和高效液相分析(例如，用于检测对映体选择性)。

Low molecular weight models for metalloenzymes are usually designed based on information available for the spatial arrangement of functional groups, metal ions, co-enzymes etc. in active sites (ideally such information is obtained via X-ray crystal analyses). This approach largely neglects the influence of the protein matrices, in particular their dynamics, local dipole moments, the inhibition of metal complex aggregation etc. As a consequence, many biomimetic low molecular weight transition metal complexes are able to reproduce structural and spectroscopical aspects of the corresponding enzymes, but most of them fail to show significant catalytic activity. In our approach we also deduce, from a given enzyme structure, which ligands are bound to the catalytically active metal center. However, in the next step, combinatorial ligand libraries are synthesized which place the metal-coordinating groups at different positions on an oligopeptide. By suitable assays, catalytically active members of the library are identified. In other words, the system is given a significant degree of freedom both in terms of constitution and dynamics, and only those entities that are catalytically active are examined further. The current project aims at the identification of functional models for galactose oxidase, i.e. biomimetic copper complexes that are able to oxidize alcohols with oxygen to the corresponding carbonyl compounds. The ligand libraries will be prepared on solid phase, encoding will be done by the IRORI radiofrequency labelling technique. The design of the libraries is based on the X-ray crystal structure of the fungal enzyme from Dactylium dendroides: histidine, tyrosine and non-natural amino acids mimicking the Cys228-Tyr272 conjugate of the wild-type enzyme will be used as metal binding entities. The non-natural Cys-Tyr mimics will be prepared in enantiomerically pure form from appropriate chirons. Assays for GOase-activity will be developed based on existing enzymatic methods for H2O2-detection, or on dye-forming reactions of the product carbonyl compounds. These reactivity assays, which are amenable to high-throughput screening, will be complemented by "classical" (chiral) GC- and HPLC-analysis (e.g. for the detection of enantioselectivity).