Next, we asked if the system can be generalized.
By varying the N-terminal residue, we created Z-XisoK tripeptides and evolved transporters for otherwise impermeable Z ncAAs, using GCE as readout for their delivery!
Z-XisoKs even enable co-delivery and co-encoding of two distinct ncAAs! 8/9

However, in nutrient-rich media (like 2-YT), uptake was less efficient – tryptic peptides present in such media competed for OppA binding.
So we evolved OppA to prefer our substrates.
Through FACS screening, we found OppA-iso and made the E. coli strain IsoK12, which thrives in complex media 💪 7/9

This revealed how the transporter recognizes and delivers our substrates.
What started as ‘unwanted cleavage’ turned into a transport system we could hijack.
Opp imports G-XisoKs, peptidases remove G, accumulating high concentrations of XisoKs for efficient incorporation via aaRS/tRNA pairs. 5/9

We discovered that E. coli actively imports G-XisoK peptides via the Opp ABC transporter that shuttles small peptides into cells in an ATP-driven manner.
We mapped the uptake mechanism, identified the peptidases removing the N-terminal G and solved the OppA:G-SisoK structure with Michael Groll. 4/9

Building on previous work from our lab on ubiquitin-protein conjugates, we aimed to incorporate G-XisoK ncAAs.
Unexpectedly, the glycine was consistently cleaved off, leaving XisoK efficiently incorporated into proteins via GCE.
A setback at first - we needed G-XisoK modified proteins. 2/9