Umwandlung einer Chorismatmutase in eine Isochorismat-Pyruvat-Lyase durch Kombination von rationalem Protein-Design und gerichteter Evolution - Erforschung der natürlichen Entstehung moderner hochselektiver Biokatalysatoren
Final Report Abstract
Enzymes are highly efficient catalysts evolved by nature to perform the chemistry of live. Their activi‐ ties and specificities enable the synthesis of complex molecules, facilitating a fine balance of all mo‐ lecules needed in cells and organisms. Adapting the biocatalysis concept to specific challenges in chemistry, pharmacy, agriculture, and energy may pave the way for improved sustainability and envi‐ ronmental protection while supplying the world’s increasing needs for energy, food, drugs and con‐ sumables. However, achieving this goal requires the design of enzymes in the laboratory for a wide variety of chemical reactions not found in nature. The failure of our efforts to convert a chorismate mutase (CM) into an isochorismate pyruvate lyase (IPL) illustrate the severe limitations on our ability to engineer catalytically active proteins and clearly highlight how little we know about the molecular origins of enzymatic efficiency. Both AroQ‐type enzymes share similar folds, reaction mechanisms, and catalytic centers, but CMs are devoid of IPL activity. When mutating the active site environment of a hyperthermostable and a molten globular chorismate mutase to compare stability and plasticity with regard to the engi‐ neering of enzyme function we were surprised that only IPL deficient proteins were generated. In order to unveil the link between protein sequence and IPL activity we subsequently challenged a comprehensive spectrum of state‐of‐the‐art enzyme engineering technologies for the modification of three CMs with varying protein stability. However, neither structure‐based design nor mathematical algorithms to decipher functional correlations between residues from protein sequence alignments were successful. Interestingly, theoretical pH titration simulations performed by the collaborating research group of Prof. Mary Ondrechen at Northeastern University (Boston/MA, USA) identified IPL residues that improved the parent activity and stability of a mesostable CM, although they also failed to confer IPL activity on this scaffold. Molecular dynamics simulations are currently being performed to investigate the molecular origins of these observations. Furthermore, the enzyme‐catalyzed IPL reaction is being studied by combined quantum mechanics/molecular mechanics calculations in the research group of Prof. Wilfred van Gunsteren at ETH Zurich. In order to enable combinatorial mutagenesis studies in high‐throughput format we devel‐ oped screening and selection assays for IPL activity. While indirect systems based on transcription activation of reporter genes by binding of the IPL product salicylate to a transcription activator failed to discriminate active and inactive library variants, fluorescence detection of salicylate in 96‐well format proved to be a robust and reliable screen. The sensitivity of this assay in the low activity range could be significantly increased by overexpression of the enzyme isochorismate synthase, which pro‐ vides the substrate for the IPL reaction and therefore enables detection of in vivo activity. However, intensive screening of randomly mutagenized libraries did not afford active lyases. After the failure of rational and combinatorial approaches we wondered why we did not suc‐ ceed in a venture that has been accomplished in nature before – the conversion of a chorismate mu‐ tase into an isochorismate pyruvate lyase – and whether activity engineering would be more success‐ ful if native evolution could be mimicked more authentically. In nature, rather than placing single variants under adaptive pressure as is commonly performed in the laboratory, new phenotypes are selected from large populations that arise from genetic drift in the absence of selection for new func‐ tion. We will use the IPL engineering problem to test whether the neutral drift concept can facilitate the generation of catalytic activity where other established methodologies failed. If successful, this study will allow improved adaptation of native molecular evolution in the laboratory and may there‐ fore pave the way for more powerful enzyme design strategies.
Publications
- Protein design by directed evolution. Annu. Rev. Biophys. 2008, 37, 153‐173
C. Jäckel, P. Kast, D. Hilvert
- Stabilizing proteins by library‐based consensus design, Poster presentation at the 22nd Symposium of The Protein Society, San Diego/CA, USA, July 19‐23, 2008
C. Jäckel, D. Hilvert
- Consensus design without phylogenetic bias, Poster presentation at the VIII European Symposium of The Protein Society, Zurich/Switzerland, June 14‐18, 2009
C. Jäckel, D. Hilvert
- Biocatalysts by evolution. Curr. Opin. Biotechnol. 2010, 21, 753‐759
C. Jäckel, D. Hilvert
- Consensus design without phylogenetic bias, Talk at the 21st Faltertage, Regensburg/Germany, October 15‐17, 2010
C. Jäckel
- Consensus protein design without phylogenetic bias. J. Mol. Biol. 2010, 399, 541‐546
C. Jäckel, J. D. Bloom, P. Kast, F. H. Arnold, D. Hilvert
- Development, characterization, and application of a recombinant antigen for the serological investigation of feline hemotropic Mycoplasma infections. Clin. Vaccine Immunol. 2010, 17, 1917‐1925
G. Wolf‐Jäckel, C. Jäckel, K. Museux, K. Hoelzle, S. Tasker, H. Lutz, R. Hofmann‐Lehmann
