Breaking Down Plastic: Researchers Develop New Enzyme with AI

In a breakthrough discovery, researchers have developed a new enzyme that can break down polyurethane, a common plastic, using AI-powered protein design tools.

TL;DR
• Researchers develop a new enzyme to break down polyurethane using AI-powered protein design tools.
• The enzyme is 450 times more active than the best-performing natural enzyme.
• This discovery could be a game-changer for plastic waste management.

Summary
Plastic pollution is a complex problem due to the various types of polymers used, each with distinct chemical bonds. Breaking down one type of polymer may not be compatible with the chemistry of another. Researchers have made progress in finding enzymes that break down common plastics like polyesters and PET, but these are only partial solutions. A new enzyme has been developed to break down polyurethane, a polymer used in foam cushioning and other applications. This enzyme is compatible with an industrial-style recycling process that breaks down polyurethane into its basic building blocks, which can be used to form fresh polyurethane.

The development of this enzyme involved testing existing enzymes reported in the literature, and then using a tool called Pythia-Pocket to identify related proteins. A neural network called GRASE was used to evaluate proteins and recommend active and stable enzymes. The results were impressive, with the best-designed enzyme showing 30 times the activity of the best-known enzyme. When combined with diethylene glycol and heated to 50° C, the new enzyme was over 450 times as active, breaking down 98 percent of the polyurethane in a reaction mixture within 12 hours.

The researchers' approach incorporated information about the enzyme's function, including stability and amino acids likely to interact with the material being digested. Their tools may provide insights into obtaining functional proteins by focusing on forming similar 3D structures. Key takeaways include:
• The new enzyme breaks down polyurethane into its basic building blocks.
• The enzyme is compatible with an industrial-style recycling process.
• The development process used machine learning tools to identify and optimize enzymes.
• The approach may be applicable to other types of plastic pollution.

The implications of this research are significant, as it could help address the complex problem of plastic pollution. By developing more effective methods for breaking down plastics, we can reduce waste and promote sustainability. This innovation has the potential to contribute to a more circular economy, where materials are continually cycled back into production.

This discovery has the potential to revolutionize the way we manage plastic waste, and it's a great example of how AI can be used to tackle complex environmental challenges.