Evaluation of 244,000 synthetic sequences reveals design principles to optimize translation in Escherichia coli

Abstract

Control of protein biosynthesis is at the heart of resource allocation and cell adaptation to fluctuating environments. One gene's translation often occurs at the expense of another's, resulting in global energetic and fitness trade-offs during differential expression of various functions. Patterns of ribosome utilization-as controlled by initiation, elongation and release rates-are central to this balance. To disentangle their respective determinants and physiological impacts, we complemented measurements of protein production with highly parallelized quantifications of transcripts' abundance and decay, ribosome loading and cellular growth rate for 244,000 precisely designed sequence variants of an otherwise standard reporter. We find highly constrained, non-monotonic relationships between measured phenotypes. We show that fitness defects derive either from protein overproduction, with efficient translation initiation and heavy ribosome flows; or from unproductive ribosome sequestration by highly structured, slowly initiated and overly stabilized transcripts. These observations demonstrate physiological impacts of key sequence features in natural and designed transcripts.

Absolute quantification of translational regulation and burden using combined sequencing approaches

“…For example, a metabolically efficient way to strongly express a protein of interest in bacteria is by producing high numbers of transcripts (e.g., with high transcription initiation rate and high stability) with a relatively weak RBS (e.g., low translation initiation rate). This ensures that each ribosome initiating on a transcript has a very low probability of colliding with others, guaranteeing efficient translation elongation (Cambray et al , ; Gorochowski & Ellis, ). We observe that this strategy is adopted for strongly expressed endogenous genes (Fig A).…”

Multiplex transcriptional characterizations across diverse bacterial species using cell‐free systems

“…Transcriptional profiles of entire genetic circuits could also be characterized in cell-free reactions using whole-transcriptome library preparation methods, which have been employed in vivo to aid debugging of individual components and dynamic circuit behaviors (Gorochowski et al, 2017). Sequencing-based quantification methods for translation such as Ribo-seq (Li et al, 2014) or polysome profiling (Cambray et al, 2018) could enable in vitro multiplex characterization of protein synthesis in different bacteria. Finally, we expect multiplexed sequencing and quantification approaches to also improve analysis of fungal (Hodgman & Jewett, 2013), plant (Sawasaki et al, 2007), insect (Ezure et al, 2006), mammalian (Mikami et al, 2006;Martin et al, 2017), and other cell-free expression systems derived from eukaryotic organisms to study and engineer increasingly complex layers of gene regulation using synthetic biology.…”

Cell-free gene regulatory network engineering with synthetic transcription factors

“…Ultimately understanding the outcome of multiple base changes in close context with each other remains a complex issue. Evaluating a greater number of sequences and systematically addressing all factors a↵ecting transcription e ciency similar to the approach taken by Cambray et al towards translation could lead to an improved understanding of promoter sequence design principles (Cambray et al 2018).…”