Evolution can be harnessed to optimize synthetic biology designs. A prominent example is recombinant protein production-a dominating theme in biotechnology for more than three decades. Typically, a protein coding sequence (cds) is recombined with genetic elements, such as promoters, ribosome binding sites and terminators, which control expression in a cell factory. A major bottleneck during production is translational initiation. Previously we identified more effective translation initiation regions (TIRs) by creating sequence libraries and then selecting for a TIR that drives high-level expression-an example of synthetic evolution. However, manual screening limits the ability to assay expression levels of all putative sequences in the libraries. Here we have solved this bottleneck by designing a collection of translational coupling devices based on a RNA secondary structure. Exchange of different sequence elements in this device allows for different coupling efficiencies, therefore giving the devices a tunable nature. Sandwiching these devices between the cds and an antibiotic selection marker that functions over a broad dynamic range of antibiotic concentrations adds to the tunability and allows expression levels in large clone libraries to be probed using a simple cell survival assay on the respective antibiotic. The power of the approach is demonstrated by substantially increasing production of two commercially interesting proteins, a Nanobody and an Affibody. The method is a simple and inexpensive alternative to advanced screening techniques that can be carried out in any laboratory.

Synthetic biology needs to adopt sound scientific and industry‐like standards in order to achieve its ambitious goals of efficient and accurate engineering of biological systems.

Supplementary Figure S1. Development and validation of a medium throughput HPLC method for the detection of L-serine in microbial culture supernatant. (a) Overlaid HPLC chromatogram of L-serine standards. (b) The recovery of L-serine from spiked samples (n=5). (c) Reproducability of the method (n=5) for different L-serine concentrations. (d) Calibration curve for L-serine.