Periodic Reporting for period 1 - fORPHAN (From protein sequence to function – computational and experimental de-orphanization of uncharacterized enzymes in fungi)
Periodo di rendicontazione: 2020-11-01 al 2022-10-31
Fast advances in genome sequencing and sequence processing technology leave ~30% of predicted proteins as orphans, meaning without known function or closely related enzymes. Being able to assign a function to such orphans opens avenues to select for and design powerful biocatalysts – individual enzymes, biosynthetic pathways or entire organisms.
The goal of the fORPHAN project was to develop a de-orphanizing pipeline based on computational and experimental characterization of enzymes from the fungal kingdom. Fungi are known to be prolific producers of secondary metabolites, however, the study of the underlying biosynthetic gene clusters is still a fairly young field. The use of automated search and annotation pipelines is currently limited by the small number of experimentally characterized gene clusters that can be used to train the algorithms. Therefore, we used deep, targeted database searches to identify key biosynthetic enzymes in the publicly available genomes of fungi. As first targets we focused on fungal type 3 polyketide synthases and putative chalcone isomerases to investigate their family-wide sequence-function relationships.
By the end of the funding period we have made significant progress in establishing our pipeline and have gained important insight into the substrate scope of fungal type 3 polyketide synthases. We have furthermore explored several microbial expression hosts for type 3 polyketide synthases. We will continue this research by further experimental characterization of the pre-screened enzymes and use them in novel combinatorial biosynthetic pathways. Two manuscripts related to the fORPHAN project are in preparation.
The goal of the fORPHAN project was to develop a de-orphanizing pipeline based on computational and experimental characterization of enzymes from the fungal kingdom. Fungi are known to be prolific producers of secondary metabolites, however, the study of the underlying biosynthetic gene clusters is still a fairly young field. The use of automated search and annotation pipelines is currently limited by the small number of experimentally characterized gene clusters that can be used to train the algorithms. Therefore, we used deep, targeted database searches to identify key biosynthetic enzymes in the publicly available genomes of fungi. As first targets we focused on fungal type 3 polyketide synthases and putative chalcone isomerases to investigate their family-wide sequence-function relationships.
By the end of the funding period we have made significant progress in establishing our pipeline and have gained important insight into the substrate scope of fungal type 3 polyketide synthases. We have furthermore explored several microbial expression hosts for type 3 polyketide synthases. We will continue this research by further experimental characterization of the pre-screened enzymes and use them in novel combinatorial biosynthetic pathways. Two manuscripts related to the fORPHAN project are in preparation.
The deorphanizing pipeline that we aimed to build in the fORPHAN project requires establishing crucial techniques in three stages of the workflow: the computational analysis of enzyme families, the enzyme profiling and substrate scope analysis of representative members of the enzyme families, and the engineering of microbial expression hosts for the potential biocatalysts. During the funding period, we have worked on all three stages in parallel and completed important milestones.
For the first stage - the computational analysis - we established a computational pipeline for in-depth analysis of amino acid sequences, active site residues, and genome neighborhoods for two families of fungal enzymes. This approach is easily reproducible and applicable to other enzyme families.
For one of the enzyme families - the type 3 polyketide synthases – we moved forward to the second stage of the workflow and performed enzyme activity essays to probe the substrate scope for more than 30 enzymes. We found that most enzymes are active on various fatty acid substrates as previously seen for other members of the enzyme family. However, we also identified a few that accepted aromatic substrates similar to plant type 3 polyketide synthases. We are currently exploring this further by scaling up the reactions and characterizing the products in more detail.
For the third stage of the workflow, we have benchmarked the expression of type 3 polyketide synthases in heterologous hosts with a few fungal and two plant enzymes. For the plant enzymes, we explored the possibility to widen the substrate scope by enzyme engineering and found that certain mutations in the active site and in a surface-exposed loop adjacent to the active site, can allow the production of the methylated flavonoids homoeriodictoyl and hesperitin (Ref. 1). We have furthermore explored the possibility to use Penicillium rubens as an expression host and found that 80% of the fed substrate p-coumaric acid is converted to the flavonoid naringenin (manuscript in preparation). While this indicates that P. rubens could be a good host for expressing biosynthetic pathways involving type 3 polyketide synthases, we also observed that it can very quickly degrade flavonoids. Therefore, we are now investigating how this competing degradation pathway can be eliminated and are also employing other fungal expression hosts such as Aspergillus oryzae.
1. Peng B, Zhang L, He S, Oerlemans R, Quax WJ, Groves MR & Haslinger K (2022) Engineering a plant polyketide synthase for the biosynthesis of methylated flavonoids. preprint on bioRxiv doi: 10.1101/2022.10.02.510496
For the first stage - the computational analysis - we established a computational pipeline for in-depth analysis of amino acid sequences, active site residues, and genome neighborhoods for two families of fungal enzymes. This approach is easily reproducible and applicable to other enzyme families.
For one of the enzyme families - the type 3 polyketide synthases – we moved forward to the second stage of the workflow and performed enzyme activity essays to probe the substrate scope for more than 30 enzymes. We found that most enzymes are active on various fatty acid substrates as previously seen for other members of the enzyme family. However, we also identified a few that accepted aromatic substrates similar to plant type 3 polyketide synthases. We are currently exploring this further by scaling up the reactions and characterizing the products in more detail.
For the third stage of the workflow, we have benchmarked the expression of type 3 polyketide synthases in heterologous hosts with a few fungal and two plant enzymes. For the plant enzymes, we explored the possibility to widen the substrate scope by enzyme engineering and found that certain mutations in the active site and in a surface-exposed loop adjacent to the active site, can allow the production of the methylated flavonoids homoeriodictoyl and hesperitin (Ref. 1). We have furthermore explored the possibility to use Penicillium rubens as an expression host and found that 80% of the fed substrate p-coumaric acid is converted to the flavonoid naringenin (manuscript in preparation). While this indicates that P. rubens could be a good host for expressing biosynthetic pathways involving type 3 polyketide synthases, we also observed that it can very quickly degrade flavonoids. Therefore, we are now investigating how this competing degradation pathway can be eliminated and are also employing other fungal expression hosts such as Aspergillus oryzae.
1. Peng B, Zhang L, He S, Oerlemans R, Quax WJ, Groves MR & Haslinger K (2022) Engineering a plant polyketide synthase for the biosynthesis of methylated flavonoids. preprint on bioRxiv doi: 10.1101/2022.10.02.510496
Based on our preliminary data, we have already identified several subgroups of fungal type 3 polyketide synthases with distinct catalytic capabilities. Therefore, I expect that we will be able to derive fundamental rules for the sequence-function relationships in this enzyme family. Several of these enzymes will be interesting biocatalysts for biotechnological applications, especially as parts of combinatorial biosynthetic pathways.
The computational workflow developed for this enzyme family, will be transferable to other classes of enzymes and allow us to derive rules for the faster prediction of enzyme function from sequences.
The computational workflow developed for this enzyme family, will be transferable to other classes of enzymes and allow us to derive rules for the faster prediction of enzyme function from sequences.