The amino acid transporter B0AT1 (SLC6A19) has recently been identified as a possible target to treat type 2 diabetes, phenylketonuria and related disorders1,2. For surface expression and catalytical activity, B0AT1 requires co-expression of collectrin (TMEM27) or angiotensin converting enzyme 2 (ACE2). Due to this interaction, B0AT1 has been recalcitrant to crystallographic analysis, and our understanding of the mechanism of its action and inhibition is limited. Its bacterial ancestor LeuT has long been used as a structural model for SLC6 transporters and is highly homologous to B0AT1 in structure and function3. They are both sodium symporters, however, LeuT is not as promiscuous as B0AT1 in its amino acid transport profile. While B0AT1 transports all neutral amino acids to some extent, LeuT only transports selected hydrophobic ones, such as leucine and alanine. Thus, it also does not recapitulate the pharmacological properties of B0AT1.
Here, we engineered LeuT to mimic human B0AT1 by mutating key residues around the primary sodium and substrate pocket, based on sequence alignment, structural alignment of various homology models of B0AT1, and previous docking and molecular dynamics studies.
The final LeuB0AT1 mutant imitates the substrate specificity of B0AT1, and is selectively inhibited by the latest generation of B0AT1 inhibitors, while WT LeuT is not. These compounds show very similar potency and mode of inhibition on this engineered bacterial system as that seen in human B0AT1. Preliminary structural studies reveal the key residues important in switching LeuT over to B0AT1-like activity in both its substrate profile and pharmacology. Together, this study sheds light on the principles of action of B0AT1 inhibitors, allowing prediction of potentially more potent and specific inhibitors.