Synthetic 5’-UTRs for Fine-tunable and Predictable Gene Expression in Escherichia coli

Abstract

Microbial engineering requires accurate information about cellular metabolic networks and a set of molecular tools that can be predictably applied to the efficient redesign of such networks. Recent advances in the field of metabolic engineering and synthetic biology, particularly the development of molecular tools as synthetic regulators in the static and dynamic control of gene expression, have increased our ability to efficiently balance the expression of genes in various biological systems.Precise prediction of prokaryotic translation efficiency can provide valuable information for optimizing bacterial metabolism for the production of biochemical compounds or engineering the bacterial host for the production of recombinant proteins. However, without carefully considering primary as well as secondary structural information of 5’-unstranslated region (5’-UTR) sequence in mRNA, it is insufficient to precisely design and modulate the gene expression level. Although many attempts have been made to develop a precise prediction tool, dynamic changes in mRNA folding throughout the translation process make it difficult to assess the folding energy determining translation efficiency and to find the exact region for standby of 30S subunit during the recruitment in translation initiation.Here, the effect of the regulatory elements such as primary and secondary structures of mRNA on the expression level was systematically investigated. Specifically, mRNA folding regions were examined to identify universal standby region of 30S subunit in Escherichia coli by correlating the folding free energies with the expression levels. It was shown that there was a critical distance from the initiation codon to undergo the translation process when 30S subunit attaches on the structured mRNAs. By assessing the exquisite regions for mRNA folding, an improved biophysical model, UTR Designer, was developed to predict the translation efficiency and demonstrated that the proper codon optimization around the 5’-proximal coding sequence is required to achieve a broad dynamic range of expression levels. Finally, the feasibility of using the developed method was discussed to control the threshold value of the input signal switching on a synthetic genetic circuit and amplify the metabolic flux toward the desired product. This should increase our understanding of the processes underlying gene expression and provide an efficient design principle for optimizing various biological systems, thereby potentially facilitating future efforts in metabolic engineering and synthetic biology.