Maintaining Plasmid Stability in Commensal Escherichia coli Nissle 1917 without Antibiotic Selection through Orthogonal Toxin-Antitoxin Systems

The rapid progress observed within the field of synthetic biology over the last decade has facilitated the design and creation of more clinically relevant and complex genetic circuits. The human microbiota consists of the various microorganisms associated with the human body and it is increasingly being implicated as a regulator of health and disease. The commensal nature of the intestinal microbiota and its constituents provide a number of well-tolerated microorganisms such as Escherichia coli NISSLE 1917 (EcN) that with synthetic biology could potentially exploit the intestines as an entry point and be used for monitoring or engineering the host microbiota. Most of the circuits designed thus far are based on the use of bacterial plasmids. These small circular DNA molecules replicate independently of the chromosome and are easy to edit and propagate with common molecular tools such as restriction enzymes and antibiotic selection markers. However, the use of antibiotic selection in the circuits would eventually hinder the investigative power of these approaches as the native microbiota would be severely impacted upon antibiotic administration. Toxin-Antitoxin (TA) post-segregational killing systems are a naturally occurring mechanism developed by bacteria to maintain plasmid stability.  This mechanism consists of a self-regulated fragment that ensures only daughter cells that possess the plasmid survive after a cell undergoes division. Here we show that an orthogonal TA system can be used in EcN to maintain the stability of a reporter plasmid in the absence of antibiotic selection for up to 13 daily passages. We characterised three TA systems from both gram-negative and gram-positive bacteria using a plasmid-born GFP fluorescent marker, flow cytometry (FCM) analysis, mathematical modelling and Bayesian statistics. Interestingly, the Axe/Txe TA system from the gram-positive Enterococcus faecium was found to be more effective in EcN than the more widely characterised and used Hok/Sok system derived from E.coli. Our results demonstrate that TA systems can be successfully used to maintain the stability of synthetic genetic circuits in commensal strains like EcN without the need for antibiotic selection. This approach provides a foundation for the construction of synthetic circuits that perform precise in vivo investigations of the intestinal microbiota within a variety of commensal bacterial strains. Synthetic biological tools like this can help further elucidate the complex role of the microbiota in human health and disease.