2022 Annual Meeting

Improving the Modularity of Bicistronic Ribosome Binding Sites for Predictable Protein Expression in Diverse Contexts

Ribosome binding sites (RBS) regulate translation initiation rates and are important tools for tuning protein expression in vivo. However, they are strongly influenced by species and local sequence. One method to reduce these effects for predictable genetic circuit behavior incorporates a bicistronic design (BCD). In this scheme, the ribosome binds to mRNA at one RBS, translates a short cistron into a leader peptide, and then initiates translation of a downstream cistron of interest using a second RBS at the end of the first cistron. By translating the first peptide, the ribosome unfolds mRNA secondary structure obstructing the second RBS, producing consistent levels of translation initiation regardless of the initial translated region (ITR) sequence. For BCDs to be widely adopted by synthetic biologists, further improvement of their reliability in different contexts is desired.

In this project, three changes were made to the existing BCD architecture. First, the distance between the first cistron’s stop and the second cistron’s start codons was increased to allow for NATG cloning overhangs. Second, the canonical GGA sequence was removed from the RBS to generate weaker variants and provide multiple options for tuning protein expression. Third, the genetic insulator RiboJ, a self-cleaving ribozyme, was inserted upstream of the first ribosome binding site to normalize transcript stability by minimizing the effects of the 5’ untranslated region (5’-UTR). BCD-containing plasmids were built using Golden Gate Assembly and cloned in Escherichia coli, while protein expression levels were measured using fluorescent protein reporters. This new series of BCDs demonstrated consistent levels of expression across a 25,000-fold range in six different ITR contexts.

We further tested the modularity of the BCDs. We showed that randomly changing nucleotides in the vicinity of the second RBS had a small (<30%) impact on expression. We also varied the distance between the codons from 0 (“overlapping”) to 5 bases. While overlapping codons produced increased protein expression levels by >50%, further increasing the gap between codons did not conclusively impact expression. All together, these results show that the new series of BCDs are broadly usable for experiments in diverse genetic contexts.