Biosensor-guided improvements in salicylate production by recombinant Escherichia coli

Abstract Background Salicylate can be biosynthesized from the common metabolic intermediate shikimate and has found applications in pharmaceuticals and in the bioplastics industry. While much metabolic engineering work focused on the shikimate pathway has led to the biosynthesis of a variety of aromatic compounds, little is known about how the relative expression levels of pathway components influence salicylate biosynthesis. Furthermore, some host strain gene deletions that improve salicylate production may be impossible to predict. Here, a salicylate-responsive transcription factor was used to optimize the expression levels of shikimate/salicylate pathway genes in recombinant E. coli, and to screen a chromosomal transposon insertion library for improved salicylate production. Results A high-throughput colony screen was first developed based on a previously designed salicylate-responsive variant of the E. coli AraC regulatory protein (“AraC-SA”). Next, a combinatorial library was constructed comprising a series of ribosome binding site sequences corresponding to a range of predicted protein translation initiation rates, for each of six pathway genes (> 38,000 strain candidates). Screening for improved salicylate production allowed for the rapid identification of optimal gene expression patterns, conferring up to 123% improved production of salicylate in shake-flask culture. Finally, transposon mutagenesis and screening revealed that deletion of rnd (encoding RNase D) from the host chromosome further improved salicylate production by 27%. Conclusions These results demonstrate the effectiveness of the salicylate sensor-based screening platform to rapidly identify beneficial gene expression patterns and gene knockout targets for improving production. Such customized high-throughput tools complement other cell factory engineering strategies. This approach can be generalized for the production of other shikimate-derived compounds.

Abstract licence: CC BY-NC-ND

A person counts once a day per paper. We keep no cookie and no address: a reader is a hash that changes every day. Any month under five reads shows no number.