Unusual Building Blocks and Domain Organization of Non-Ribosomal Peptide Synthetases
The diverse class of non-ribosomal peptides consists of manifold pharmacologically important natural products. They are clinically used in antibiotic, antiviral and antitumor therapy, furthermore some are known immunosuppresants. The biological activity is based on their structural diversity, as the...
|Online Access:||PDF Full Text|
No Tags, Be the first to tag this record!
|Summary:||The diverse class of non-ribosomal peptides consists of manifold pharmacologically important natural products. They are clinically used in antibiotic, antiviral and antitumor therapy, furthermore some are known immunosuppresants. The biological activity is based on their structural diversity, as they contain various non-proteinogenic building blocks and amino acids of which many are beta-modified. It was shown that the latter are important for biological activity, but little is known about their biosynthetic origin. In particular, these building blocks are key determinants of the class of acidic lipopeptide antibiotics and kutznerides, which are in the focus of this thesis. To determine the mechanism underlying the biosynthetic origin of the synthetically challenging beta-hydroxylated asparagine (hAsn) moieties, found in the acidic lipopeptides CDA and A54145, the corresponding recombinant non-heme iron (II)/alpha-ketoglutarate dependent hydroxylases AsnO and LptL have been examined in vitro. Direct hydroxylation of the free amino acid was observed in both cases, clearly indicating a precursor synthesis pathway. The crystal structure of one of the two hydroxylases (AsnO) was determined at high resolution and revealed a substrate induced fit mechanism of the enzyme. Upon addition of asparagine, a lid-like region seals the active site and shields it from sterically demanding substrates, which explains the observed specificity for free asparagine. Furthermore, the AsnO structure could be seen as an archetype enzyme for non-heme iron hydroxylases acting on free amino acids. It was possible to predict amino acid binding residues for homologous enzymes by 3D modeling. In order to fully understand the mechanisms of beta-hydroxylated building blocks synthesis, the hydroxylases KtzO and KtzP, predicted to be responsible for the generation of the two 3-hydroxyglutamic acid isomers found in the mixture of antifungal and antimicrobial kutznerides, were produced recombinantly and analyzed in vitro. Notably, they were found to work in trans to the assembly line on PCP-tethered glutamic acid rather than on the free amino acid. Unexpectedly, as the two isomers are found in approximately equal amounts in mature kutznerides, KtzO was shown to stereospecifically generate threo-hydroxyglutamate, while KtzP catalyzed the formation of the erythro isomer by co-elution HPLC experiments with synthetic dabsylated standards. A powerful method that employs non-hydrolyzable coenzyme A analogs was developed, which allowed the determination of the kinetic parameters of enzymes working on PCP-bound substrates for the first time. Furthermore, a hitherto unknown mechanism of NRPS assembly line restoration was observed. The corresponding adenylation (A) domain for glutamic acid activation in the kutzneride NRPS was found to be corrupted. Herein, it is shown that this lack of a functional A domain is compensated in trans by a stand-alone A domain. These findings elucidated the mechanism for the in trans compensation and the stereospecific hydroxyglutamate generation in detail and may guide the usage of in trans hydroxylation/compensation enzymes in biocombinatorial engineering approaches. In the third part of this work, the acquired knowledge about the mechanisms underlying enzymatic beta-hydroxylation of amino acids was exploited for the synthesis of the pharmaceutically relevant beta-hydroxyaspartate. Primarily, this was facilitated by the structure elucidation of AsnO in which the substrate binding residues were identified. By site directed mutagenesis, an AsnO variant was generated, which notably did not hydroxylate the original substrate asparagine, instead it was found to stereospecifically catalyze the formation of L-threo-hydroxyaspartic acid, even in commercially interesting amounts. Therefore, the AsnO variant is an excellent example for the application of basic research in order to generate pharmacologically relevant non-proteinogenic amino acids.|