PLP-dependent enzymes: from genomics to chemical compounds

Amino acids are among the most important molecules in nature and chemistry since they play central roles both as intermediates in metabolism and as building blocks of proteins and pharmaceuticals. In living organisms enzymes that depend on pyridoxal-5-phosphate (PLP) catalyze a wide variety of biochemical reactions involving amino acids substrates and their analogs. These enzymes have been exploited by chemists for the biocatalytic synthesis of natural amino acids, and most recently have been investigated for the asymmetric synthesis of non-canonical amino acids. During the project we will explore and broaden the catalytic potential of PLP-dependent lyases, in particular threonine aldolases, for the application as biocatalysts to produce - and -substituted -amino acids. The high specificity towards natural substrates and sometimes low stereoselectivity is the limiting factor for the use of the PLP- dependent lyases in organic synthesis. Rational design represents an efficient approach for engineering of enzymes with desired properties. However, identification of functional residues is challenging, especially when structure-function relationships in an enzyme is poorly understood. Looking back to the evolution of enzymes within one family of PLP-dependent enzymes is essential to understand how functionalities were evolved by Nature. The evolution based knowledge will be used to design or evolve new substrate specificities and catalytic activities in the enzymes of interest. During the research on PLP-dependent enzymes, their genetic and structural information will be compared using modern bioinformatic tools in order to: a) increase knowledge on structure-function relationships in PLP-dependent enzymes and identify the amino acid residues responsible for functional discrimination between enzymes sharing common evolutionary origin; b) predict mutationsnecessary for improvingsubstrate (e.g. donor) specificity and diastereoselectivity of threonine aldolases and design catalysts for the synthesis of -hydroxy- quaternary--amino acids; c) expose the catalytic potential of PLP-dependent lyases with versatile catalytic properties (e.g. tryptophan synthase, tyrosine phenol-lyase and cystathionine -lyase) and evolve their promiscuous activities for the formation of -substituted -amino acids; Rationally designed and evolved lyases will be applied for the asymmetric synthesis of non-canonical amino acids using a novel biocatalytic approach with high selectivity under mild reaction conditions, which will be a promising alternative towards currently used organic chemistry methods.

The use of biotransformations for the synthesis of chemical compounds is an efficient and sustainable alternative to classical synthetic methods. In this project, we aimed to investigate the synthetic potential of the natural catalysts - enzymes, for the synthesis of amino acids. Amino acids are among the most important molecules in nature and chemistry; they play central roles both as intermediates in metabolism and as building blocks of proteins and pharmaceuticals. In living organisms enzymes that depend on pyridoxal-5'-phosphate (PLP) catalyze a wide variety of biochemical reactions involving amino acids substrates and their analogs. These enzymes have been exploited by chemists for the biocatalytic synthesis of natural and non-natural amino acids. However, their application in industry is still limited due to narrow substrate range and sometimes low activity toward molecules of interest. During the project, we explored and broaden the knowledge on the PLP-dependent lyases, in particular, threonine aldolases, for the application as biocatalysts to produce beta-hydroxy-alpha-amino acids. This knowledge allowed us to rationally design wild-type L-threonine aldolase and create the biocatalysts with desired properties. Consequently, L-phenylserine derivatives, potential drugs for Parkinson's disease therapy, were efficiently produced using simple one-step biocatalytic transformation with engineered L-threonine aldolase. Further, alpha-substituted analogs of L-phenylalanine, important building blocks of therapeutic peptides, can be produced with excellent stereo-purity using chemo-enzymatic two-step procedure. In addition, upon a comparison of evolutionary peculiarities in the PLP-dependent enzymes, we discovered how the functional divergence was created by natural evolution. This finding enables chemists to create new specific biocatalysts and synthetic pathways but also gives essential knowledge to pharmaceutical scientists to discover new drugs candidates.