There’s a bit of a dirty little secret in synthetic biology – gene circuits that work great in one species often require a lot of changes to work in anything else. This is one reason so many synbio papers contain the same two key words: E. and coli. Now, a new approach called UBER (Universal Bacterial Expression Resource) promises to make it a lot easier to get where you want to go in a strange new environment (no app yet, though).
The main problem in porting gene circuits between species is that expression levels change. Different bugs express genes from different promoters and ribosome binding sites at different rates. One solution is to use an orthogonal RNA polymerase, like that from the T7 phage, which recognizes a unique promoter. Unfortunately, T7 RNAP at high levels can be toxic to cells.
In their new UBER system, Manish Kushwaha and Howard Salis devised a set of coupled feedback loops to keep T7 RNAP at a desired level. In the positive feedback loop, T7 RNAP produces itself from its own promoter, and in the negative feedback loop, it also transcribes tetR, which represses T7 RNAP production. For translation, they designed a library of ribosome binding sites (RBSes) that should work similarly across species.
The UBER system for drag-and-drop gene circuits. Image: Manish Kushwaha & Howard Salis
This feedback-controlled T7 RNAP was able to produce high GFP output while avoiding host toxicity. When tested in three species (E. coli, P. putida, and B. subtilis), the system had a wide range of outputs, but at least different RBS variants had the same output rank from low to high in each species.
The team also did some rigorous modeling, and in an interesting side note, they found that competition for T7 RNAP among promoters has a big impact on output gene expression. When tetR represses the T7 RNAP promoter, that frees up more RNAP to bind the output GFP promoter, resulting in more output even with less RNAP.
UBER is a good step toward making drag-and-drop gene circuits that work the same in any species you drop them into. True, the output spread from B. subtilis to E. coli was ~5-10-fold for an identical circuit, but that’s not so bad, considering that the copy number varied from 1 chromosomal integration to 50 plasmids. Gene expression can be adjusted with known libraries of T7 promoters and UBER’s universal RBS library, making it much easier to work with species that don’t have standard genetic parts.
Hail yourself an UBER over at Nature Communications, July 2015