Can I use multiple bicistronic RBS sequences in a synthetic biological circuit?

Question

The bicistroninc RBS sequences (BCDs) developed by Mutalik et al. [1] aim to remove context sensitivity from translation and therefore ensure more predictable gene expression. However, I have been told that having multiple transcriptional units, each using BCDs, can make cells sick. The explanation that I was given was that the toxicity is from small peptides that are produced from transcription from the first RBS sequence.

Has anyone demonstrated toxicity from multiple BCDs or is it apocryphal? If it has been demonstrated, is the mechanism of toxicity known and is there a way to use multiple BCDs while maintaining healthy cells?

[1] V. K. Mutalik et al., “Precise and reliable gene expression via standard transcription and translation initiation elements,” Nat. Methods, vol. 10, no. 4, pp. 354–360, Apr. 2013.

Answer 1

That’s a great question and has lot of opportunities to explore. I am not sure anyone has followed up on this original BCD work systematically. We did try cloning these elements on a medium copy plasmids, in an operon design, driving multiple genes, and now I think about it, that was not a smart thing to do. Cloning was very challenging, probably due to toxicity you mention. Didn’t have time to investigate this in depth (that it self is a cool project), but I think the toxicity is due to too much resources (ribosomes?) titrated out on these units. But that’s a hand-wavy explanation. Need mechanistic study to investigate. I think Tom Ellis’s group and others have done some beautiful work on toxicity associated with protein expression. Personally I think the toxicity is not related to the inert peptide (1st gene in a BCD operon) that’s produced. Another story we tried was to replace this first gene with something functional but not published.

Coming back to using BCDs for driving multiple genes, I would say if you clone these directly on genomes they should work like charm. We did try this and works good. Not published much (oh god, that’s so much work unpublished!) Bacterial operons have these junctions and elements of different strengths. Happy to discuss more if you have a follow up question!!

Thanks for your interest in this work. Really proud of our work on this system. Of course happy that we did this work as there were so many early effort on these BCD type elements. And love this open discussion SE platform! Don’t know how synbio survived without this system of sharing knowledge and experiences. Best wishes

Vivek Mutalik

Answer 2

In the documentation for this recent cloning method from Richard Murray's group at Caltech (,https://doi.org/10.1021/acssynbio.8b00060) they include a parts library with various BCDs. I haven't tested this directly myself, but the warning they note for multiple BCDs is not that they are toxic, but that because they have long homologous regions that they are prone to recombination. This makes sense since the dummy peptide region is I think around 50bp and shared across all the BCDs. You could imagine making variants in this region to combat recombination issues, but who knows if that would affect the robustness of the BCD expression.