Model-based enzyme evolution – MoBEE
Directed evolution is nowadays the most efficient approach to design enzymes with improved or new properties. This experimental method subjects a candidate sequence to cycles of diversification and selection that mimic the natural evolutionary process. In practice, however, it requires to start with a sequence only a few mutations away from the sequence of interest. Our project aims at quickly accessing by this method regions of the sequence space very distant from the starting point, which is crucial for the evolution of enzymes with new functions. As a demonstration, we propose to generate variants of an enzyme with wild-type activity but mutations on more than 50% of their sequences, using just a few selection cycles in combination with new data analysis techniques.
To this end, we will develop mathematical models that combine biophysical insights and statistical inference. The models will be fed and validated with high-throughput sequencing data generated by a new experimental system for high-throughput selection. Our strategy is based on a well-established relationship between the biophysics and the evolutionary properties of proteins: proteins with increased thermal stability are more robust to mutations and more ‘evolvable’. We will learn how thermal stability is encoded into sequences by analyzing the data that our experiments will generate, where sequences are selected for catalytic activity. From this inference, we will design libraries of sequences with enhanced stability, which we will subject to new rounds of evolution. This will generate data to improve the model and navigate further in sequence space. Importantly, our directed evolution experiments will be performed in vitro and using a self-selection protocol, which will produce extremely high-throughput data at low cost and with minimal biological bias.
Our project should lead to new fundamental insights into the relationship between the sequences of enzymes, their stability and their catalytic properties, as well as to new practical approaches to stabilize enzymes and evolve them towards new catalytic activities.
Project coordination
Olivier Rivoire (GULLIVER)
The author of this summary is the project coordinator, who is responsible for the content of this summary. The ANR declines any responsibility as for its contents.
Partnership
CIRB Centre interdisciplinaire de recherche en biologie
GULLIVER
GULLIVER
GULLIVER GULLIVER
Help of the ANR 399,716 euros
Beginning and duration of the scientific project:
- 48 Months