Living organisms are fantastic chemists. They are able to use enzymes during chemical reactions by using the gentlest conditions.
Creating brand-new enzymes to accelerate chemical reactions is a crucial part in the field of protein design with massive implications across different fields. For instance, in the field of biotechnology, enzymes can increase biofuel production, food processing, and pharmaceutical manufacturing. They can also be used as therapeutic and diagnostic agents. For environmental remediation, enzymes can help in breaking down pollutants and assist in the production of biodegradable materials.
A team, composed of two postdoctoral scholars, based at the Institute for Protein Design at the University of Washington Medicine utilized deep-learning, artificial intelligence and came up with machine-learning algorithms that created light-emitting enzymes called luciferases. In the creation of new luciferase enzymes, the team first selected luciferins that they wanted the proteins to act on. Using software, they were able to produce thousands of possible protein structures that might react with luciferins.
The researchers were able to identify an efficient enzyme that performed the desired chemical reaction named LuxSit, short for “Let there be light”. The enzyme was further refined which only led to pronounced improvements in performance. The optimized enzyme, called LuzSit-i, produced enough light that could be visible to the naked eye. Compared to the natural luciferase enzyme found in Renilla reniformis, its light was considered to be brighter.
The laboratory testing confirmed that the new enzymes could efficiently identify specific chemicals and emit light. In principle, this innovation led to the realization of the ability to design custom enzymes for almost any chemical reaction.