CASE STUDY — Drug Biomanufacturing

Overview

Transaminases catalyze a reaction between an amino acid and an alpha-keto acid, making them a key ingredient in creating molecules for a wide range of industrial and medical applications.

Challenge

While more than 70 substrate-specific transaminases have been identified, there is always a need to find better options that can do more under a variety of challenging conditions, especially those capable of working with bulky substrates which are of paramount importance to the pharmaceutical industry. Notably, while the majority of natural enzymes sourced from bacteria exhibit S-selectivity in chirality, only a limited number of R-selective transaminases have been identified to date. To truly augment the applicability of transaminases for biocatalysis, further identification of these R-selective enzymes becomes imperative.

Structure-based enzyme discovery and variant generation tools 

Starting with existing R-selective transaminases as reference, we conducted sequence-based and structure-based enzyme discovery. This combined search leveraging bit-QED informatics allowed us to expand the search space for identifying potential R-selective transaminases from our proprietary microbial genome database, bit-GEM. 

Utilizing the natural TA sequences sourced from bit-GEM, we applied ancestral reconstruction modeling to generate variants with enhanced properties including increased thermostability, solvency in organic solvents, and salt tolerance.