Computational Identification and Characterization of Putative Baeyer–Villiger Monooxygenases from Tenebrio molitor Microbiomes for Future Low-Density Polyethylene Degradation

• Rowan Moxley, Chemical Engineering, University of Delaware

• Erik Breiling, Chemical and Biomedical Engineering, University of Delaware

As the search for plastic-degrading methods continues, the demand for green chemistry and cost-effective plastic treatment strategies remains high. One potential approach to plastic recycling is Baeyer–Villiger oxidation, which has been shown to degrade plastics such as low-density polyethylene (LDPE). Baeyer–Villiger monooxygenases (BVMOs) may catalyze similar reactions on LDPE through an enzymatic pathway rather than a purely chemical process. Such an approach could provide a safer alternative to conventional chemical treatments while enabling more cost-effective LDPE recycling. However, BVMOs have not yet been evaluated for their ability to degrade LDPE, partly because relatively few BVMOs have been well characterized. To enable future studies of BVMO reactivity toward LDPE, we first sought to identify and characterize putative BVMOs using computational approaches. Our search focused on the microbiomes of plastic-degrading yellow mealworms (Tenebrio molitor), allowing for a more targeted enzyme discovery strategy. Candidate enzyme selection used one computational model, while NADPH depletion assays were used as a high-confidence experimental method to evaluate BVMO activity. Computational enzyme identification combined machine learning with amino acid sequence analysis, using conserved motifs to narrow candidate selection without requiring three-dimensional structural information. Using this workflow, we identified 22 putative BVMOs computationally. Four candidate enzymes were selected for experimental validation and tested against their predicted substrates using NADPH depletion assays. Putative BVMOs 1, 3, 5, and 6 all exhibited NADPH consumption at varying rates across their respective substrates, supporting the computational model’s ability to identify enzymatically active BVMO candidates. The four enzymes then were tested with Gas Chromatography to verify the products and better characterize their activities. Overall, these tests allow for BVMOs to be better understood and will segway into streamlined testing of possible LDPE degrading abilities of such enzymes.