(567l) Same Needle, Smaller Haystack: Computationally Designed Degenerate Ribosome Binding Sites to Identify Optimal Enzyme Expression Levels

Synthetic metabolic pathways can convert common cellular metabolites into a wide variety of valuable chemicals, including fuels, drugs, and materials. A central goal is to identify the optimal enzyme expression levels that maximize the pathway's productivity or titer. One approach is to insert a degenerate ribosome binding site (RBS) sequence to randomly vary the expression levels of the proteins of interest. The best combinations of protein expression levels are selected by conducting a functional high-throughput screen. However, this approach scales poorly as the number of proteins is increased. For example, a degenerate RBS containing 6 Ns will contain 4096 sequences and selecting the near-optimal expression levels of only 3 proteins will require a library of 6.9e10 sequences.

We present a computational optimization algorithm that designs degenerate ribosome binding site sequences to systematically and efficiently identify optimal enzyme expression levels. The algorithm has two objectives: maximize the expression level coverage of the degenerate RBS sequence, while minimizing the number of sequences. We use the recently developed RBS Calculator (http://salis.psu.edu/software) to predict the translation initiation rates of degenerate RBS sequences and to calculate how well the library of sequences uniformly samples the protein's expression level across a 100,000-fold range (the coverage). The multi-objective optimization algorithm identifies a specific degenerate RBS sequence that achieves these goals. We experimentally validate the optimization algorithm's predictions by measuring protein expression levels in a synthetic test system. We then demonstrate how using designed degenerate RBS sequences can identify the optimal protein expression levels for a metabolic engineering application.

The design method for degenerate RBS sequences enables the efficient identification of the optimal protein expression level, and can be applied towards controlling the expression levels of transcription factors, transporters, or enzymes for metabolic engineering applications.