CRISPR-based curing and analysis of metabolic burden of cryptic plasmids in Escherichia coli Nissle 1917

E. coli Nissle 1917 (EcN) has long been used as an over-the-counter probiotic and has shown potential to be used as a live biotherapeutic. It contains two stably replicating cryptic plasmids, pMUT1, and pMUT2, the function of which is unclear but the presence of which may increase the metabolic burden on the cell, particularly in the context of added recombinant plasmids. In this work, we present a clustered regularly interspaced short palindromic repeats-Cas9-based method of curing cryptic plasmids, producing strains cured of one or both plasmids. We then assayed heterologous protein production from three different recombinant plasmids in wild-type and cured EcN derivatives and found that production of reporter proteins was not significantly different across strains. In addition, we replaced pMUT2 with an engineered version containing an inserted antibiotic resistance reporter gene and demonstrated that the engineered plasmid was stable over 90 generations without selection. These findings have broad implications for the curing of cryptic plasmids and for stable heterologous expression of proteins in this host. Specifically, curing of cryptic plasmids may not be necessary for optimal heterologous expression in this host.