Researchers have used artificial intelligence (AI) to design brand-new enzymes that can efficiently undergo multi-step reactions, a key feature of natural enzymes. This breakthrough could have significant implications for biological and industrial processes, including plastic recycling.
The development of these new enzymes is a major milestone in enzyme engineering, as it shows that AI can be used to create enzymes with native-like activity. The researchers combined several machine-learning approaches to overcome the challenge of designing enzymes that stall after the first step of a reaction.
The traditional approach of tweaking existing enzymes to create new ones has limitations, as it makes it difficult to create efficient enzymes capable of multi-step reactions. In contrast, the AI-driven design process allows for precise control over the enzyme’s structure and function.
The researchers aimed to create enzymes that can undergo a four-step chemical reaction called serine hydrolysis, which is involved in various biological processes such as digestion and blood clotting. The new enzymes were 60,000 times better at speeding up the reaction than previously designed ones.
This achievement marks a significant innovation in enzyme engineering, as it demonstrates the potential of AI to create functional enzymes that can perform complex tasks.
Source: https://www.nature.com/articles/d41586-025-00488-3