Although the selection of new enzymes was largely based on activity detection in fungal cultures, and already sequenced genomes, two new genomes of fungi scarcely investigated (the ascomycetes Lecytophora hoffmanii and Kretzschmaria deusta) were sequenced in the project (2,3). Cultures and genomes were screened for new types of unspecific peroxygenases (UPO), the most promising oxidoreductase type in EnzOx2 (4,5), whose industrial applicability was improved by different enzyme engineering methods and structural-functional studies (6-10). Oxidases were also investigated to catalyze reactions of interest or as a source of the hydrogen peroxide required by UPOs. Efficient in situ generation of H2O2 has been addressed in several EnzOx2 studies (11-13). Aryl-alcohol oxidase (AAO) was selected as a model oxidase, and its reaction mechanism with both reducing and oxidizing substrates was investigated (14-16) for enzymatic catalyst. The above applications include new enzymatic technologies for both sugar and lipid derived building blocks and added value compounds.
Concerning sugar (furfural) based building blocks, the project focused on the enzymatic conversion of HMF to 2,5-furandicarboxylic acid (FDCA), a renewable precursor for the production of poly(ethylene-2,5-furandicarboxylate) (PEF), the polymer expected to substitute for petroleum-derived poly(ethylene-terephthalate) (PET) plastics in the near future. We optimized the previous two-enzyme cascade for HMF conversion into FDCA, and proposed a new cascade involving AAO, UPO and galactose oxidase with additional interest (17). Interestingly, a recent study revealed that the inability of AAO to complete HMF conversion into FDCA is, at least partially, due to inhibition of the last oxidation step by the H2O2 generated during the two previous oxidation steps and, therefore, can be abolished by catalase (18).
Since large-scale production of FDCA-based bioplastics (e.g. by the joint venture of BASF and Avantium) will be based on 5-methoxymethylfurfural (MMF), this compound was also included in EnzOx2. A self-sustained three-member cascade for FDCA production from MMF was developed (13) and patented (19), where the action of UPO is fueled by hydrogen peroxide provided by both AAO and a second oxidase acting on the methanol release from the demethoxylation of MMF by UPO. Finally, the production of PEF from enzymatically-synthesized FDCA was evaluated by AIMPLAS and Ava Biochem.
For the enzymatic production of lipid "sensu lato" added value compounds, the action of UPOs on different compound types has been investigated. Among them, the enzyme from Marasmius rotula catalyzes unique reactions on fatty acids including terminal/subterminal hydroxylation and chain shortening. The latter has been fully investigated from enzymatic and mechanistic points of view (20) and a patent has been deposited for the controlled one-carbon shortening of fatty acids (21), a reaction representing a novel chemistry that may be used in biotechnological applications to obtain tailor-made acids not abundant in nature.Terminal hydroxylation, of interest for homopolyester production and other applications, is being investigated with UPOs.
Among steroid transformations, the enzyme from Marasmius rotula (and a second Marasmius species) was also able to remove the side chain of cortisone for production of APIs (22) in a reaction reminiscent of the chain shortening mentioned above. A second steroid transformation of pharmaceutical interest was demonstrated for the UPO of the ascomycete Chaetomium globosum, which catalyzes the selective oxygenation of testosterone (23). This reaction, yielding 90% testosterone epoxide (isolated with 96% purity) has been recently up-scaled and positively evaluated for industrial implementation. Two additional reactions of pharmaceutical interest catalyzed by several UPOs are the hydroxylation of stilbene (and several stilbenoids) for the selective synthesis of resveratrol analogues (24) and the hydroxylation of propranolol (25,26).
Challenging selective oxyfunctionalization of four model terpenes has been actively investigated by the above and other UPOs, although the developed bioprocesses did not fulfil the feasibility requirements for their industrial implementation in the flavour & fragrance industry. Moreover, selective synthesis of 4-hydroxyisophorone and 4-ketoisophorone, of interest for both pharmaceutical and flavour & fragrance sectors, has been reported with UPOs (27) and protected by a patent application (28).
The industrial impact of the above enzymatic transformations is related to: i) the use of environmentally-friendly enzymatic technologies in the manufacture of bio-based bulk chemicals such as plastic building blocks; and ii) the selectivity of UPOs catalyzing specific oxygenation reactions of pharmaceuticals, and other speciality chemicals, that are difficult or very complicated (expensive) to be obtained using chemical methods. Several publications on their environmental feasibility have been already produced in EnzOx2 (29-31).
